Foreign direct investment and competitiveness in the Romanian manufacturing

The objective of the analysis is the static (the database refers exclusively to year 1998) evaluation of the competitiveness of Romanian manufacturing, depending on the ownership-type. A ternary ownership-type structure is considered: the prevailing foreign-owned capital, the prevailing Romanian private capital and the prevailing state capital. The place, the share, the performance and the impact of the foreign ownership in the Romanian manufacturing are described. The methodology of the analysis is subordinated to the signification given here to the concept of competitiveness: "the capacity of producing and selling profitably, both on the domestic and international market". Moreover, taking into account the large share of losses in the '98 Romanian manufacturing (approximately one third of the firms, which account for approximately half of the total of fixed assets), two sectors were defined and individually analyzed: the sector of the profitable firms and the sector of the unprofitable firms. The relation of the turnover, the exports, the productivity, the investment, the materials, etc. with the profits and the losses are considered in connection to the ownership-type. Finally, a competitiveness criterion is defined. The criterion is implemented, taking into account the ownership-type, to evaluate the Romanian manufacturing industry, as a whole and at NACE-division level. Chapter I The theoretical bacground of the relationship foreign direct investment-competitiveness I.1 Theoretical aspects The existence of a potential positive impact of foreign direct investment (FDI) on the competitiveness of host-countries is widely acknowledged by investing companies, authorities of recipient economies, analysts. Presenting the coordinates of such an impact is the subject of a quite large number of studies, which sustain with concrete examples the positive effects induced by FDI through: the transfer of a complex package of resources (capital, technology, management and marketing skills); opening access to new markets; inducing microeconomic restructuring; generating spillover effects for local partners, suppliers or clients of the foreign affiliate, or in the rest of the economy; increased revenues to the state budget; and boosting exports. Disagreement appears as soon as the word potential is left aside. Those who unconditionally place themselves on neo-liberal positions will plead for the effectiveness of a positive impact irrespective of conditions. The anti-globalists will deny a priori such an impact, while, on a third position, are to be found those analysts who considers that assessments should be made only by a case-by-case approach. According to this third opinion, there are no generally accepted conclusions concerning the gains of the host-countries from FDI. These will depend on the concrete situation determined, on one hand, by the capabilities, interests and strategy of the investing firm and, on the other hand, by the development stage, the characteristics of the economic environment and the policies of the recipient state. This paper addresses the specific aspect of FDI-competitiveness linkage under the major objective of Romania in the following years, that is joining the EU. Or, in its economic dimension, this process mostly depends on the speed at which the Romanian economy will be restructured and become more competitive. As shown before, foreign capital can play an important role in the respective process. We intend to evaluate the current effects of FDI on the competitiveness of the Romanian economy both at microeconomic level, that is at recipient firms level, and at macroeconomic level, that is their restructuring impact through a better allocation of resources. The theoretical framework in which the restructuring role of FDI is more recently and frequently grounded is the competitive advantage theory of Michael Porter (Porter, 1992). Porter considers that the increased mobility of production factors and the major role of innovation in all respects (especially in what information technologies are concerned) severely limit the explanatory capacity of the neoclassical theory of international trade and the validity of its recommendation that countries should specialize according to their production factors endowment. Being competitive has become less a problem of maximizing results within a given environment and more and more a dynamic challenge for states and firms to generate innovation, to create new factors and to improve the existing ones. Porter designs his own structure of determinants of the competitive advantage of nations, at industry level. He considers competitiveness to reside mainly in the productivity with which a nation makes use of its resources in a certain economic activity. The competitive advantage are determined by that level of productivity that allows local firms to generate substantial and sustained exports to a significant number of countries or important FDI flows. At this point, we would like to stress a fact that derives from Porter's work and also from Dunning's important contributions to the study of international production and FDI effects in host-economies: not only the generation of FDI, but also, even if at a different scale, receiving FDI is a measure of the competitiveness of an economy. Porter's theory supports the idea that, despite the globalization of production and trade, the competitive advantage is created in a national framework, nations, through their institutional, natural, cultural, economic characteristics ultimately determining the development of certain economic activities. The factors considered by Porter as determinants for the competitive advantage are grouped in four categories, the linkages between them being important as well. The system as a whole has a diamond type structure which also gives the name of Porter's model. The four categories of decisive factors for creating and sustaining national competitive advantages are the following (Porter, 1992, p.69-131): F1. production factor endowment: human resources, both under a quantitative and qualitative aspect; natural resources; technical, scientific, market knowledge; capitalavailability, cost, structure of the financial market; physical and business infrastructure. Given the present sophistication of production organization, the direct access to production factors no more represents a condition of competitiveness. Factor mobility, the prime role of created factors, especially technology, a sustained innovative process of creating new factors are presently such conditions; F2. home demand characteristics, that is its level and structure, its degree of sophistication, its capacity of formulating anticipative needs. Home market expectations put pressure on the sellers, determining them to improve their offer, while demand anticipative requirements, anticipative as against other markets, can make home products competitive, through exports and FDI; F3. the level of development of related and supporting industries. The competitive advantage in a certain industry is also determined by the performance of supplying or beneficiary industries; F4. firms structures and strategies and home rivalry. This forth category of determinants is linked to the managerial culture as a factor which can influence firm organization and way of operation and as a potential source of competitive advantage. A managerial culture, which encourages a high degree of moral commitment of the members of an organization, reduces the need for mutual surveillance and encourages cooperation among the members of the organization. While a set of principles which stresses respect of the authority at the cost of respect of the others will enhance the efficiency of hierarchies. Managerial culture is expressed by firm organization and way of operation but most of the firms don't create the culture, they inherit it. The British author Mark Casson (Casson, 1990, p.88-94), distinguishes between: (i) the technical aspects of a culture, which include a scientific outlook and influence the individuals' and firms' perception of the environment and hence the quality of the decision-making process in both the economic and technological spheres and (ii) the moral values of the society which influence economic performance by legitimating certain general principles of behavior and also by encouraging entrepreneurial commitments and determining the intensity with which individuals strive to honor these commitments. Some moral attitudes are far more entrepreneurial than others. The philosophy of voluntarism, for example, which legitimates the freedom of business is more favorable to the development of the entrepreneurial spirit than a value system which concentrates the coercive powers on institutions such as the state. A competitive climate is decisive for creating and maintaining the competitive advantages of firms, because they are the economic actors whose performance on international markets substantiates the competitive advantages of a nation. Local rivalry is more visible and more direct than the international one and at least as important as this one. In function of the main source of the competitive advantage at a certain moment, Porter defines the stages of development of a nation. In the presentation of these stages and of the role FDI can play we'll also take into account the contributions of John Dunning (1993, p.272-276) and of the Japanese author Terutomo Ozawa (Ozawa, 1992, apud Dunning, 1993, p.272). S1. The stage of the production factor driven competitive advantage. According to Porter's model, the first development stage is the production factor driven competitive advantage. He considers that the great majority of developing countries as well as the former centrally-planned economies from Eastern and Central Europe were in this stage at the beginning of the '90s. More than this, countries such as Canada and Australia were, according to the main source of competitive advantages, in the same first stage of development. Dunning shows that in this phase, FDI will be directed towards the primary sector and towards low-qualified labor-intensive manufacturing activities. It is a stage in which created production factors are scarce and hence their contribution in the economy is a modest one. In terms of policies, the states in this phase tend not to impose restrictions or performance criteria on FDI inflows. It is true that the lower the level of high national capabilities, the greater is the risk of enclave creation through FDI in the host-countries. S2. The stage of the investment driven competitive advantage. In this phase, the competitive advantages are mainly determined by the volume and quality of investments in modern technologies and production facilities, in developing a competitive physical and business infrastructure. According to Dunning, in this second stage of development, domestic investment share in gross domestic product (GDP) may rise from 5-8% to 15-20% (Dunning, 1993, p. 273). At the same time, there is an increase of expenditure on secondary education, public utilities, transport and communications. In this development stage, the sources of competitive advantages are likely to shift towards capital-intensive sectors, such as basic chemicals, iron and steel and ship-building; some smaller scale mechanical engineering activities; and the production of labor-intensive, but moderately knowledge-intensive consumer goods, such as electrical products, clothing, leather goods, processed foods and cigarettes. The role of foreign capital will mainly depend on the development strategy promoted by the recipient countries governments. The options of the Japanese and South-Korean authorities in the '50s were to restrict the amount of inward investment and develop their own asset capabilities. On the contrary, starting with the '60s, Ireland integrated FDI into its industrial development policy, succeeding in improving its economic performance, including the sharp increase of exports of electronic and pharmaceutical products. FDI inflows can play the role of initially stirring certain economic activities, generating "virtuous" circles of asset accumulation. The development of connection activities may also be sustained through FDI infusion or –on the contrary, which is the situation to be preferred– by local firms. S3. The stage of the innovation driven competitive advantage. Within Porter's model, the intensive investment phase is followed by the stage in which the competitive advantage of a nation mainly derives from its capacity to innovate. It is the phase in which the technological and managerial progress and the updating of the production facilities are mainly undertaken by indigenous firms. Local firms are now capable to compete in more and more specialized and narrow market segments, on the basis of global strategies, under the positive impact of a sophisticated local demand. Activities that rely on traditional production factors are transferred beyond the national borders. Most of the developed states are or have already passed over this stage. The localization of the innovative process doesn't mean that FDI can no more play a role in the development of host – economies. The increased mobility of competitive advantages in today world economy allows firms to develop specific ownership advantages which can be successfully used in innovation stage economies. S4. As far as the next development stage is concerned, Porter's opinion differs from that of Dunning. Porter's model includes as the last stage of the development cycle the welfare phase. Welfare by itself is likely to cancel the motivation for sustained investment and innovation. Competitive advantages are more and more volatile and the return to the first stage of development, based on production factors and their costs become necessary. Porter considers that this is the case of Great Britain in which, at the beginning of the '90s, the low cost of the labor force played such a role. At its turn, Italy ran through the whole cycle, presently covering the innovation – led development stage. In Dunning investment cycle path, the final and most advanced stage of economic development is the information processing stage, also called the post – industrial or services stage of development (Dunning, 1993, p. 274). Nevertheless, the author admits that countries like US, Japan, Norway and Germany, which according to his model are in this last stage, also meet the most significant conditions that place them in the innovation stage. They remain the leading spenders on research and development activities, which, for the most part, are directed towards the innovation of new products and production methods. There should be also mentioned the fact that given the radical technological advances in computing and telecommunications, the traditional borders between manufacturing industries and services are likely to disappear. A proof in this respect is the increased content in services of material goods. In our opinion, all these indicate the possibility of considering the information processing stage as a superior phase of the innovation – led stage in Porter's model. Irrespective of its condition as a distinctive phase or an integral part of the third development stage in Porter's theory, the development of information processes strongly increases the number of linkages among companies. The success of countries in accumulating productive assets is depending more and more on local ability to coordinate the use of resources in a regional and even global environment. In this phase, the effects of FDI inflows in recipient countries are to be valued not only in terms of resource transfer, but mainly in terms of organizational practices and skills which can benefit by means of spillover effects the abilities and efficiency of local firms. The sequence of development stages doesn't apply as such to all states; Dunning doesn't exclude the possibility for certain states to straddle more than one stage at a time. More than this, within a country, different regions may be at different stages of development. Porter's model and Dunning's considerations theoretically substantiate the approach to FDI – competitiveness relation in Romania. The assumption we start from – assumption that we intend to evaluate as true or not – is that Romania is at the first stage of development, stage in which its competitive advantages are determined by production factor endowment and by the low cost of the respective factors. According to Porter and Dunning, in this stage, FDI are oriented towards primary production processes and towards manufacturing activities based on the low cost of the labor force, being able to play an important role in enhancing the economic performance and in improving the production factors quality. The centrally planned economies disposed to a very little extent, if at all, of the mechanics allowing for the creation of specialized production factors. The lack of domestic competition, the restriction or even the total call off of the demand role in the economy – not to speak of a possible pressure put on suppliers by a sophisticated demand –, the allocation of resources on non-economic criteria, all these led to competitive advantages based exclusively on factor cost and located in standardized market segments. As shown before, according to Michael Porter, transition economies are at the first stage of development of their competitive advantages. His opinion is shared by East-European analysts like Ferenç Vissi, once the president of the Competition Office in Hungary who brings about some new aspects, correlating the sequence of development stages with the degree of exposure of a country to the competition on international markets (Chikán and others, 1998, p. 12-13). While analyzing the place and role of foreign trade in Romania's economic development, the economist Valentin Cojanu approaches the problem of the competitive advantage stage of the Romanian economy. His conclusion is in conformity with the above-mentioned assumption. Romania, a country with economy in transition was in the first half of '90, at the first stage of development of its competitive advantage, stage based on the production factors endowment and their low costs. The author considers that, unfortunately, the actual potential of the Romanian economy doesn't allow for "its rapid development and advance towards superior stages" (based on investment and innovation) (Valentin Cojanu, 1997, p.208). I.2 Approach directions Our paper approaches the relationship between FDI and competitiveness only in the manufacturing industry. The reasons for this limitation reside in the fact that the restructuring impact of FDI flows as well as their spillover effects in the rest of the economy depend, to a great extent, on the volume and quality of the FDI flows in manufacturing. Four main directions of analysis, D1 – D4 will be approached D1. Our analysis starts with a secondary, but still necessary step with respect to the objectives of the paper: defining the place, the share of the foreign sector in the Romanian manufacturing, at a general level, as well at NACE subsection level. By this, we intend we examine the validity of Dunning assertion that in the first development stage, FDI flows are mainly directed to low valueadded manufacturing activities, which turn into account the low cost of labor force. D2. The second natural coordinate of the analysis refers to FDI impact on the economic performance of recipient firms, the point from which every debate concerning FDI effects should start from. Given the fact that the great majority of FDI in Romania have been made in existing companies or within the privatization process and not in the form of greenfield investments, a complete analysis should be a dynamic one. This would solve the problem of causality, discriminating between 1) FDI impact on economic performance of recipient firms –FDI impact on performance subsequently the FDI infusion– and 2) the degree of attraction of local firms by means of their economic performance, for foreign investors, which influence the investment decision before the infusion of FDI. A correct approach of the subject should include the examination of the performance of the same company before and after receiving foreign capital. Unfortunately, some problems make difficult if not impossible for the time being such an approach: changing the ownership structure involves more than one stage (in Romania, joined ownership has still got an important share); there is an objective transition period after the infusion of foreign capital, transition whose impact on firms' performance cannot be neglected. Such restrictions would require a larger time span, which in Romania's case doesn't exist: the opening of the economy has only started at the beginning of the '90s. An alternative solution would be a case-by-case approach, which is not what we intend to do. Most of the analysis carried on by east-European economists, including Romanian ones, are based on comparing the performance of companies with foreign capital with the average industry performance, or with the performance of local firms. To a great extent, this aspect was covered (Boşcaiu and others, 2000). Some of the conclusions of this study, relevant for the problems of competitiveness are to be interpreted and used in our paper too. D3. At a third tier, we intend to deepen the analysis of competitiveness at the level of NACE divisions in manufacturing. The starting point consists in the observation that the condition of factors F1-F4 is reflected in the profitability of manufacturing and at firm level in the value of profits or in that of losses. In this respect, the analysis of profits and losses is an important part of the competitiveness analysis. In the analysis, we'll not explicitly approach the four categories of competitiveness determinants (factors F1-F4 mentioned above), but we'll analyze in detail the profits and losses according to the ownership type. D4. Finally, we intend to determine the FDI role in the efficient allocation of economic resources within manufacturing, inclusively by comparison with the allocation carried on by companies with prevailing domestic capital. Specific indices will be used, among which the allocation diversity ratio. 1.3 Methodological comments M1. Defining the ownership structure of the company. The first specification to be made refers to the ownership structure of the company. As far as the binary model "companies with foreign participation versus domestic companies" is not valid because of the lack of homogeneity of local capital, a ternary approach will be followed: -prevailing state-owned companies (abbreviated STATE), -prevailing private-owned domestic companies (abbreviated PrivRO), -prevailing foreign-owned companies (abbreviated FOR). These abbreviations will be systematically used in the paper (although licences according to the literary expression standards, abbreviations have the advantage of facilitating reading). The term "prevailing" is used to identify the main source of the capital. Theoretically, prevailing foreign-owned companies (or, similarly, "prevailing state", or "prevailing domestic private") means the possession of at least 34% of the equity capital, but, in most cases, (in Romania), it is equivalent to the majority capital (more exactly, if the equity capital belongs to a certain ownership form in proportion of over 34%, than, in most cases, the respective ownership form represents over 50% of the equity capital). This approach is necessary to make possible the covering, within the analysis, of the companies with a joined ownership structure whose frequency in the Romanian manufacturing industry was in 1998 (and still remains) important. M2. Description of the sample. Initially, there have been taken into account all the approximatively 2900 firms in manufacturing which, in 1997 and 1998, had an average number of employees of at least 50 people. Subsequently, there were excluded approximatively 120 firms from various reasons: incomplete, incorrect information, outliers or because in 1998 the weight of the trade activities in the turnover was of over 50% (although, according to the main object of activity declared, the firms belonged to manufacturing). Therefor, finally, a sample of 2800 firms was accepted for the analysis. In order to ensure the homogeneity of the paper and the possibility of subsequent computing of other indices, starting from those computed within the study, all the aggregations were made for this sample, even if for some indices the available data would have permitted to use a larger sample. The initial intention was that the sample would be exhaustive, but finally, it wasn't (because of nonresponses, incomplete, incorrect or untypical data). Thus, the sample is neither exhaustive nore random. This doesn't raise major problems given the fact that, at each subsection level, our data base covers at least 80% of the turnover. The conclusions of our analysis can be, hence, considered as representative for the Romanian manufacturing as a whole. In what the subsection analysis is concerned, the results must be interpreted separately, in connection with the number of firms in each subsection (see Annex 2, Table 4). M3. The "sheep louse" firms, which parasite the Romanian industry were automatically removed from the data base, through the two selection criteria: firms with at least 50 employees and a weight of trade activities less than 50%. M4. The data base at firm level which is used in this paper refers to the situation at the end of 1998. The analysis is nevertheless valid for the next two years as well, given the fact that FDI flows in 1999-2000 registered similar levels: 818.5 million dollars in 1998, 930.6 million dollars in 1999 and 865.1 million dollars in 2000 (The Trade and Industry Chamber of Romania, 2001. Chapter II The size and performance of the foreign sector in the Romanian manufacturing industry II.1 The size of the foreign sector in the Romanian manufacturing industry We will evaluate the share of the foreign sector FOR by examining he values of some parameters calculated in function of the three ownership types mentioned above. The following aspects will be taken into consideration: ownership structure of the capital, turnover, exports, investments, valueadded, profits, number of employees. Each firm is classified as belonging to a certain NACE subsection according to the main object of activity (for the manufacturing industry, from EA, food, beverage and tobacco industry to EO, other industrial activities, see Table A1). The analysis refers to the situation at the end of 1998, but as shown before we consider it actual and valid for the present as well. Fixed asset distribution. At the manufacturing industry level, the FOR companies account for 9.5% of the total fixed assets, against 29%, the PrivRO companies and 61.5%, the STATE companies. It is obvious that at the end of 1998, the share of FDI in the Romanian manufacturing industry was a low one. Nevertheless, at the level of NACE subsections, there are some branches in which the share of the predominantly foreign capital-FOR is higher: EM, electrical and optical equipment industry (27%); EI, industry of other products made of non-metallic minerals (20%); EG, chemical and synthetic and artificial fibers industry (18%); EA, food, beverage and tobacco industry (17,7%); EB, textile industry and of textile fabrics (17%). Turnover. The shares of the three ownership forms in the aggregate turnover for the manufacturing industry were at the end of 1998 the following: 47% the STATE companies, 39% the PrivRO companies and 14% the FOR companies. The highest shares of FOR were to be found in: EM, electrical and optical equipment industry (34.8%); EC, leather and footwear industry (28.6%); EA, food, beverage and tobacco industry (26.4%); EG, chemical and synthetic and artificial fibers industry (26%); EB, textile industry and of textile fabrics (22.1%). The lowest shares of FOR companies in the subsection turnover were in EF, EJ, EH, EL, and respectively EN as it follows: EF, industry of oil processing, coal coking and nuclear fuels processing 0%; EJ, metallurgical industry – in 1998, only 0.1%, but the share increased afterwards; EH, rubber and plastics manufacturing industry 4.1%; EL, machine and equipment building industry4.5% and, finally, EN, industry of transport vehicles 5.6%. Exports. As regards the shares of FOR companies in exports of the subsections, the respective companies account for over 20% of the turnoverin the following branches: EB, EK, EM, EG, EC, ED and EE. The percentages are the followings: 28% in the textile industry and of textile fabrics; 27% in the metal structures, metal products industry (except machines, equipment, installations); 27% in the electrical and optical equipment industry; 25% in the chemical and synthetic and artificial fibers industry; 22% in the leather and footwear industry, in the wood working industry (except furniture) and in the cellulose, paper, cardboard, paper and cardboard products industry. For the manufacturing industry as a whole, over 50% of exports are generated by the STATE companies (approximately 52% of the STATE companies, while the PrivRO and FOR firms account for 34.5% and 13.4%, respectively). Investments. The contribution of FOR companies to the investment process was in 1998 of 23% for the manufacturing industry. Shares almost twice as big were registered by the FOR sector in: EG, the chemical and synthetic and artificial fibers industry (49,6%); EC, leather and footwear industry (46,4%); EM, the electrical and optical equipment industry (44,5). The industry of other products made of non-metallic mineralsEI, the food, beverage and tobacco industryEA, as well as the wood working industry (except furniture) also witness levels of over 30% of the FOR participation to investments. Value added. The participation of the FOR companies to the value added creation in the manufacturing industry is low, of only 12.9%. The highest level of this participation, of 31.4% is registered in EG, the chemical and synthetic and artificial fibres industry and the next, of 21.8% in EA, the food, beverage and tobacco industry. Profits. The FOR sector account for approximately one quarter (24.2%) of the profits value generated in the manufacturing industry. Their share is relatively high in the following subsections: EC, leather and footwear industry (48.2%); EM, electrical and optical equipment industry (40.8%); EG, chemical and synthetic and artificial fibres industry (38.5%), EA, food, beverage and tobacco industry ( 35.3%) and EB, textile industry and of textile fabrics (32.2%). Number of employees. The wholly domestic-owned firms, STATE and PrivRO, account, each group for around 46% of the workforce in the manufacturing industry, while the foreign sector employs only 8%. The most significant shares of the FOR are to be found in EB, the textile industry and of textile fabrics, EC, leather and footwear industry and EG, the chemical and synthetic and artificial fibres industry, with levels situated between 13-15%. II.2 The economic performance of FDI The immediate theoretical background of FDI impact on the performance of recipient firms and, as a result of spillover effects, on that of wholly domestic-owned companies, as well are to be clearly found in the monopolistic advantage theory and in the theory of internalization. Both theories try to explain foreign direct investment and international production starting from the existence, at firm level, of certain competitive advantages (monopolistic or oligopolistic advantages, including technological ones, the privileged access to financing or suppliers etc.) or internalization advantages that result from operating production assets within a common ownership and organizational structure. On the basis of such advantages, foreign firms are able more than to offset the difficulties of being present into a new foreign environment, facing the local competition and proving themselves to be profitable. The argument of the efficiency of foreign firms is even more visible as valid in the case of FDI motivated by the factor costs, as in such situations, foreign markets are the ones to confirm the competitiveness of the respective products. This subchapter carries out an analysis of FOR sector performance in comparison with PrivRO and STATE sectors. We intend to continue the analysis of FDI impact on competitiveness at microeconomic level by deepening the above findings at NACE subsection level. We'll try to identify, on one hand, the industrial branches in which the FOR companies register the best performance and, on the other hand, the branches in which the respective firms have a significant contribution to certain economic indices. In this respect, we'll compare the performance of the FOR firms with the PrivRO and STATE sectors. Labor productivity. Labor productivity is calculated as a ratio between value added and number of employees. In order to complete the above analysis, we'll resort to the labor productivity ratio (LPR), defined as an aggregate index at subsection level, calculated as a ratio between labor productivity by ownership and subsection labor productivity (see Def.1, Annex 1). LPR is maximal for FOR companies in all the subsections of the manufacturing industry, with one exception, insignificant from a statistical point of view (in the case of rubber and plastics manufacturing industry). The highest levels of the LPR are registered by FOR firms in the following industries: chemical and synthetic and artificial fibers industry (237%); cellulose and paper industry (193%); food, beverage and tobacco industry (192%); electrical and optical equipment industry (183%); industry of transport vehicles (151%), as well as the machine and equipment building industry (204%, but in this case the share of the FOR sector is low). For the manufacturing industry as a whole, LPR value for FOR companies is 164%, compared to 96% in the case of the STATE firms and 93% in the case of PrivRO firms. Capital productivity. Capital productivity is calculated as the ratio between turnover and the value of fixed assets (machines, equipment and transport vehicles). The capital productivity ratio (KPR, see Def.1, Annex 1) is defined as an aggregate index at subsection level by the ratio between capital productivity by ownership type and the capital productivity for the subsection as a whole. KPR points out the superiority of FOR companies only in 4 out of the 15 subsections of the manufacturing industry: in the leather and footwear industry (112%); in the cellulose and paper industry (161%); in the chemical and synthetic and artificial fibers industry (227%); in the machine and equipment building industry (216%). For the entire manufacturing industry, the KPR of FOR companies is 125%, higher than that of Romanian STATE companies (76%), but lower than that of PrivRO companies (135%). Marginal productivity. The below conclusions refer to a Cobb-Douglas type model of the production function with four production factors (in a large sense), namely capital value, labour, value of materials and value of subcontracting: the production is defined by δ γ β α S M L AK Y = (Boscaiu and others, 2000). (Under this model, the marginal productivity of capital is αY/K, the product of α which describes the contribution of capital K in the production function and Y/K, a measure of capital productivity; similar results hold for L, M, S.) The marginal productivity of capital is significantly higher for the private firms (PrivRO or FOR) against the STATE ones, the difference between FOR and PrivRO being insignificant. The marginal productivity of the labor is higher in the PrivRO sector, without, this time, significant differences between STATE and FOR. The marginal productivity of materials (materials are described by the total cost of raw materials, materials, energy and water) is maximum for STATE companies and minimum for the FOR companies, all differences between the three ownership types being significant. These differences may be related to two aspects: export-import activities (taking into account that the share of imported materials is significantly higher for FOR firms and for the companies with regular exports) and, on the other hand, the share of value added in turnover (which is minimal for FOR firms). A final aspect revealed by the regression analysis refers to the marginal productivity of subcontracting, which is the highest for FOR companies, followed by PrivRO and STATE companies. This fact is important from the point of view of the integration of the foreign affiliated activity in the Romanian economy, subcontracting being one of the ways in which the degree of integration may rise. It is true that the other main way consists in buying raw materials and materials from the Romanian suppliers; it was shown before that as far as this criterion is concerned, FOR firms are on the last place. The analysis of the total productivity factor (TFP). We must have in mind that TFP covers all the causes that can alter productivity, others than those directly related to the production factors. TFP expresses the current level of technology as well as the efficiency of the organizational and managerial practices. In a previous study (Boscaiu and others, 2000, using a log-linear regression model) there has been found that the total productivity factor is significantly higher for the private firms (PrivRO or FOR) in comparison with the STATE companies. The difference between FOR and PrivRO is positive but not significant. Exports. Another set of conclusions refers to the frequency and intensity of export activities, which are maximal for FOR companies. More precisely, 47% of the FOR firms register an over 75% share of exports revenues in turnover, against only 18% of the PrivRO firms and 4,2% of the STATE companies (in number terms). This conclusion contradicts, at least in the case of the Romanian manufacturing industry, the quite spread perception that the main motivation of the companies with foreign participation is the domestic market. FOR companies mainly work for export in the following industries: textiles, metallurgical industry, machines and equipment building, transport vehicles and furniture. On the other hand, there are industries FOR companies work almost exclusively for the domestic market: food, beverage and tobacco industry (income from exports represents only 1% of the turnover), rubber and plastics manufacturing (10%), electrical and optical equipment(14%), cellulose and paper industry (21% of the turnover). At the aggregate level of the manufacturing industry, the highest share of export income in the turnover of 29% is generated, nevertheless, by STATE firms, against 25% in the case of FOR firms and 23.5% in that of PrivRO firms. A coherent interpretation of this apparently paradoxical situation asks to point out some aspects previously mentioned: the small numerical share of STATE companies (approximately 25%), but the high share of their turnover (approximately 60% of the aggregate turnover of the manufacturing), the aggregate distribution of exports in the manufacturing industry (52% for STATE companies, 35% for PrivRO companies and 13% for FOR companies). This means that if almost half of the FOR firms account for the great majority of exports, the propensity for export of the other half is so low that, on the whole, the foreign sector performance is below, even if not much below, that of STATE companies (25% against 29%). For all exporting companies, the percentages of the export losing firms (in the sense that revenues from exports are smaller than expenses for exports) are 20% for STATE and FOR against 10% for PrivRO. Moreover, 25% of the FOR firms with substantial exports (more exactly, which export at 1 A “higher” value of the marginal productivity means a “higher” effect in the production function (a higher output) of an added unit of the considered resource. least 75% of the production), are exports losers. In this case, too, the situation of PrivRO firms is relatively better: the percentage of the export losing firms is of approximately 10%. Imports. The higher propensity for exports of FOR firms is accompanied by their higher inclination for imports as well. The share of imported materials (raw materials and materials, except energy and water) in the total expenditures of this kind of the company is significantly higher for the FOR sector. In the textile industry as well as in the leather and footwear industry – subsections with the highest levels of imported materials – the share of imported materials is 100% for more than half of the FOR companies. It should be also mentioned that the share of imported materials is higher for the firms with systematic exports, especially for those whose exports exceed 50% of the turnover. Investments. The level of investments represents another aspect to be taken into account when judging the companies' performance. The investment intensity is expressed by the share of the investments in turnover. The investment ratio (see Def.1, Annex 1) will defined as an index aggregated at subsection level by the ratio between the investment intensity by ownership type and investment intensity of the subsection. At the level of the manufacturing industry, the investment ratio of FOR companies (163%) is more than double compared to that of STATE companies and over 50% higher that that of PrivRO companies (108%). The highest level of the ratio registered by the FOR companies is to be found in the industry of transport vehicles: 435% (against only 74% for the STATE companies and 110% for PrivRO). Other subsections in which the contribution to investments of the FOR companies is higher than that for the manufacturing industry as a whole are the industry of other products made of non-metallic minerals (315% against 17% in the case of STATE companies and 102 in the case of PrivRO companies); wood working industry, except furniture (223% against 14% and 89%, respectively); and finally, chemical and synthetic and artificial fibers industry (191% against 61% and 85%, respectively). The industry in which the propensity for investment of the FOR companies is minimum, being, at the same time, uncharacteristic for this group of firms, is the metal structures and metal products industry, except machines, equipment, installations. The level of the ratio is of 24%, compared to 18% in the case of the STATE firms and 156% in the case of PrivRO firms. The share of value added in turnover. At aggregate level, the share of value added in turnover is the lowest for the FOR. The value added ratio (expressed by the respective share for each ownership type and the share at the manufacturing industry level, see Def.1, Annex 1) is of 91%, for the FOR firms, nevertheless quite close to that registered by the STATE Romanian firms, of 93%. Both levels indicate a performance below the medium level (100%). The sector that achieves the best level of this index is the PrivRO sector, with a value of 112%. The FOR companies exceed the 100% level in the following subsections: chemical and synthetic and artificial fibers industry (121%), the industry of transport vehicles (109%) and the industry of other products made of nonmetallic minerals. II.3 Conclusions concerning the size and performance of the FOR sector The review of both the performance and share indices (see also Table A2 in the Annex 2) provides some important conclusions concerning the FOR sector. We mention them below. 1. At an aggregate level, the weight of value added in turnover is minimal in FOR sector. This finding confirms the opinion according to which the foreign direct investments in Romania are mainly located in low value added activities. At the same time, at firm level, the share of imported materials is the highest for the FOR companies. These are two of the explanations of the low macroeconomic impact of foreign capital on the restructuring processes in the Romanian economy. 2. In comparison with the domestic-controlled companies (STATE or PrivRO), the FOR firms are superior from the following points of view: their labor and capital productivity and the frequency of their export activities are higher (even if for manufacturing as a whole, the share of exports in turnover is maximum for the STATE sector); their investment effort (described as the ratio between investments value and turnover) is much higher compared to STATE and PrivRO and their management (at least in what their capacity of using subcontracting is concerned) is superior. 3. In Romania's case, there is obvious the existence of a number of superlatives of the FOR firms, compared to the companies controlled by the domestic capital. Nevertheless, on the other hand, the frequency of losses is higher for FOR than for PrivRO. As for the frequency of loss-making exports, this one is maximum for the FOR firms, also. Therefore, there is a polarized structure of the foreign-controlled companies. Thus, the profit-making firms are efficient and numerous enough to impose as a trend, at the aggregate level of the FOR sector, a number of superlatives concerning economic performance, and the unprofitable firms are numerous enough to significantly increase the frequency of losing firms. 4. The polarization of the PrivRO sector, if it exists, is much smaller than that of the FOR sector. For the PrivRO, the superlatives are fewer but the losses are smaller, too. 5. As it can be pointed out, the above conclusions doesn't offer enough indications for the evaluation of the FOR sector competitiveness in the manufacturing industry. Chapter III The analysis of the foreign capital competitiveness in manufacturing: aggregated analysis of profits and losses We'll admit the definition of competititveness as the capability of producing and selling profitable. In this context, the profits analysis, the losses analysis and the profitability analysis –the subject of this chapter – are absolutely necessary for assessing competitiveness. In this section, "the profitable sector" and "the unprofitable sector" denominate the set of all the firms that in 1998 made profits or, respectively, registered losses. The separation of the profitable firms from the unprofitable ones is explained by the existence of major, systematic discrepancies between the two categories, discrepancies that, in most cases, cannot be explained by favorable or unfavorable economic circumstances and can not considered to be random. III.1 Frequency of profitable companies, profitability In 1998, the frequency of profit-making firms was of approximately 68% for FOR compared to 76% for PrivRO (for STATE, the percentage was of only 45%). The FOR sector accounts for maximal values in 7 divisions, while PrivRO in 14 divisions. Columns 6-9 in Table A4, Annex 2 present the distribution of profitable firms at NACE divisions level. We enumerate below the divisions with statistically significant discrepancies, only (the number of firms in each analyzed category must be large enough). Without insisting upon these values, we will mention the divisions that witness major discrepancies between the three ownership types (discrepancies will be marked with "!"). The divisions with an important variability of profits frequency are the following (the codification of the divisions is made explicit in Table A1, Annex 2): Div 15 (the percentages of profit-making firms are 36%!, 69% and 58% for STATE, PrivRO and FOR respectively), Div 17 (27%!, 65%, 59%), Div 18 (33%!, 86%, 76%), Div 19 (38%!, 72%, 79%), Div 24 (31%!, 74%, 75%), Div 27 (33%! For STATE and 71% for PrivRO), Div 28 (52%, 79%, 36%!), Div 36 (32%!, 70%, 71%). We shell describe the profitability by the ratio between the difference "profits minus losses" and the turnover (percentage expression; see Table A5, Annex 2). In 1998, the profitability of the Romanian manufacturing industry was negative: -1.29%. This means that the manufacturing industry contributed to the decreasing and not to the increasing of the national wealth. The main contribution to this counter-performance was made by the STATE sector, with a profitability of -7.85%, while the private sector (domestic and foreign) had a positive contribution. The NACE divisions in which the STATE firms made profits were only three: Div 16, Div 22 and Div 32. It is interesting to notice that in the three profitable divisions, the STATE sector had a better performance compared to that of the private capital, although exports registered low values. Thus, losses were small or absent (95-100% of the assets were in the profitable area) and the share of profits was significantly higher (see Table A5). Nevertheless, on the other hand, among the industries in which the STATE firms registered losses there are some considered to locate competitive advantages, such as the textile industry, the leather and footwear industry, the wood working industry and furniture. At industrial branch level, the profitability of FOR and PrivRO is very different. The divisions in which the profitability of the FOR sector is much higher to that of the PrivRO sector are the following: Div19, the leather and footwear industry (the profitability of FOR companies is 16.38% against 1.03% for the PrivRO companies); Div20, the wood working industry (6.73%, against the negative -0.66%); Div21, the cellulose, paper and cardboard industry (11.35%, against -0,87); Div25, rubber and plastics manufacturing industry (9,31% against 0,89); Div31, the electrical and optical equipment industry (17,58%, against 7,60%). The divisions in which the profitability of the FOR sector is much lower than that of the PrivRO sectorur are the following: Div22, publishing houses, polygraphs and type copying (the profitability of the FOR firms is 3.23% against 8.16% for the PrivRO firms); Div24, the chemical and synthetic and artificial fibres industry (1.80% against 9.75%); Div 26, industry of other products made of non-metallic minerals (1.39% against de 11.26%); Div 28, the metal structures and metal products industry, except machines, equipment, installations (-53.40% again 5.24%); Div29, the machine and equipment building industry (-3.09% against de 4.24%); Div35, the industry of other transport vehicles (-35.73% against 13.10%); Div36, furniture and other non-classified activities (-3.79% against 2.54%). The divisions in which both FOR and PrivRO sectors register some of the highest profitabilities: Div18, ready-made clothes of textiles, furs and leather (12.81% and 14.60%, against de -7.91% for STAT). III.2 The location and size of the profitable sector The dimension of the profitable sector can be described by three indices. The first one, already used in §III.1, is the frequency of profitable firms. This one presents the disadvantage of equally treating every two firms, irrespective of their dimensions. The second index is the share of the profitable capital (SPA), defined by the share of the value of fixed assets of the profitable sector, in the total of fixed assets value (the share will be expressed as a percent). A third index will be used, as well, the share of the profitable turnover (SPT)-similarly defined, with reference to the share of the turnover. More precisely, the definitions of the indices are the following: SPA =100 X FAprof / FAtot SPT=100 X Tprof / Ttot where: FA= aggregate capital, expressed by the aggregate value of fixed assets in the reference set (the manufacturing industry or a NACE division); T= aggregate turnover in the reference set. 'prof ' and 'tot ' indicate the aggregation sets: the profitable firms of the considered set and all the firms of the considered set, respectively. At the level of manufacturing, the share of the profitable capital is SPA=49%. In this way, the share of the "captive" assets in the unprofitable sector is of 51%. That is – in more vague but suggestive terms – half of the Romanian manufacturing was losing in 1998. The turnover of the profitable sector was of approximately 2/3 of the aggregate turnover of manufacturing. The decreasing order of the share of the profitable sector within each ownership type is the following (see Table A3 in Annex 2): PrivRO (70% of the capital and 82% of the turnover are concentrated in the profitable sector), FOR (62% of the capital and 78% of the turnover located in the profitable sector) and STATE (with only 36% of the capital and 44% of the turnover to be found in the profitable sector). Irrespective of the ownership type, the profitable sector accounts for large shares in Div16 (accounting for 100% of the capital), Div32 (93% of the capital), Div18 and Div22 (85% of the capital), Div31 and Div26 (with 78% and 74% respectively of the capital). At division level, the share of the profitable sector generally presents a higher variability with regard to the ownership type. The divisions with a relative homogeneous behavior according to this criterion are just a few: Div16, Div22, and Div32, exactly those with the lowest share of the unprofitable sector. The shares of the profitable sectors are described in the following table. Table III.1 The distribution of the profit-making and loss-making sectors, depending on prevailing ownership and NACE division 0) Prevailing ownership Divisions locating a high share of lossmaking sector 1) Polarized divisions 1) 2) Divisions locating a high share of profit-making sector 1) STATE Div17, Div18, Div23, Div24, Div25, Div27, Div33 (>64%) Div15, Div19, Div20, Div26, Div28, Div29, Div31, Div34, Div35, Div36 Div16, Div22, Div32 (>95%); Div21 ( 61%) PrivRO Div21 (79%) Div17, Div19, Div20, Div24, Div25, Div27, Div36 Div18, Div23, Div26, Div29, Div31, Div32, Div33, Div35 (>80%); Div22 (78%); Div34(74%); Div28(68%); Div37(62%) FOR Div27, Div28 (>75%); Div29 (74%) Div25 Div16, Div18, Div19, Div21, Div32, Div34, Div36, Div37 (>80%); Div15, Div26 (73%); Div22, Div24, Div31(65%); Div17, Div20(61%) The manufacturing industry, as a whole Div23, Div27(85%) Div17, Div19, Div20, Div21, Div24, Div25, Div28, Div29, Div33, Div34, Div35, Div36, Div37 Div16, Div18, Div22, Div32, (>85%); Div26(78%), Div31(74%); Div15(64%) 0) The table summarizes information for the manufacturing industry 1998, from the Table A3(see the columns 6-9), Annex 2. 1) Classification criterion is the capital share of the division, defined by the percent of the fixed assets value of profitable firms, computed in the total value of the fixed assets. All divisions were considered, irrespective theirs (small) number of firms. 2) A division is defined as "polarized" if profit-making and loss-making sectors have near shares (in the interval 40-60%). III.3 The profit, the losses and the average size of the capital It should be noticed that the analysis of the frequency of profitable firms from §III.1 and the analysis summarized in Table III.1 are not equivalent: the conceptual difference consists in the fact that the first refers to the firm level, while the second to the aggregate level. Therefore, the frequency of the profitable firms and the share of the profitable capital are only partially consistent, although they register a high empirical correlation coefficient at division level: 0.84. Moreover, taking into account the two variables, the capital rate, RK, will be defined (see Table A4). The index RK will be computed for each division (and for the manufacturing industry as a whole, as well) and for each of the three ownership types, STATE, PrivRO and FOR (but also irrespective of the ownership type) as it follows: RK = (FAloss/FAprof)/(NRloss/NRprof) = (FAloss/NRloss)/(FAprof/NRprof) where: FA = aggregate capital, expressed by the aggregate value of fixed assets; NR = number of firms 'prof' and 'loss' indicate the aggregation set: profitable firms, unprofitable firms, respectively. In this context, FA/NR is the mean value of the capital and it describes in fact the average size of the firm, while RK compares the average sizes, more exactly it describes the ratio between the mean calculated for the loss-making firms and the mean for the profit-making firms. If the rate value is significantly higher than one, RK>1, then the average size of the loss-making firms is higher than that of the profit-making firms. (for example, RK=2 means that, in a specified set of firms, the mean capital of the loss-making firms is twice as big as the mean value the profitmaking firms). Otherwise, if RK<1, then the average size of the loss-making firm is lower than that of the profit-making firms. A value of RK that doesn't significantly differ from 1 (is approximately 1) should be interpreted as it follows: the hypothesis of equality of the mean capital of the profitable and unprofitable firms is accepted. The term "significantly" should be understood under the meaning of the theory of testing the statistical hypothesis. We draw the attention that in the absence of a correctly substantiated statistical analysis, the values of RK should be considered in the following empiric, "weaker" manner: if RK<1, then the bigger firms have a greater probability (but we don't know how much greater) of being profitable and, opposite, if RK>1 then it is "more probable" that the small firms should be profitable. At the manufacturing industry level (see the first line of Table A4), it can be noticed that for all the three ownership types, the relatively small firms present a greater probability to be profitable. This is also the case for some divisions: Div25, Div27, Div28, Div34 and Div35. Opposite, in the divisions Div15, Div18 and Div32 the bigger firms have a higher probability to be profitable and this is true for all the three ownership-types. The above suggests us the following conjecture: "the RK values can offer valuable information to underlie the restructuring policies". (Example: an increment of the size of the small firms in Div15, Div18 and Div32, could increase the probability to be profitable.) But we must stress that this conjecture will remain an hypothesis only, until a more stringent analysis will be carried on; it is obvious that the influx of capital cannot automatically make the firm profitable. We present below a synoptic table of the divisions in which the value of RK is significantly different from 1. We mention the fact that some "very high" values of RK –see Table A4– couldn't be taken into account because they were not statistically significant, mainly due to the small number of firms. Table III.2 Profitability description , depending on the capital ratio RK , prevailing ownership and NACE division Prevailing ownership Divisions with RK>1 2) Divisions with RK<1 3) STATE Div26, Div27, Div34 Div15 PrivRO Div17, Div19, Div20, Div21, Div24, Div28, Div34, Div36 Div15, Div18, Div32 FOR Div28 Div15, Div18 The manufacturing industry, as a whole Div17, Div19, Div20, Div24, Div27, Div28, Div29, Div34, Div35, Div36 Div15, Div18, Div22, Div32 0) The table summarizes information for the manufacturing industry 1998, from Table A4, Annex 2. 1) See above the definition of the index RK. 2) If RK>1 then the mean capital of the loss-making firms is significantly higher than the mean capital of profit-making firms. 3) If RK<1 then the mean capital of the loss-making firms is significantly less than the mean capital of profit-making firms. III.4 The link between profits, losses and the capital productivity Similar to defining the capital rate, we define RKPR, the capital productivity rate as the ratio between the productivity of the profitable capital and the productivity of the unprofitable capital. This index can be calculated as functions of SPT and SPA (see Table A3). RKPR is defined as it follows: RKPR = (Tloss/Tprof)/(FAloss/FAprof) = (Tloss/FAloss)/(Tprof/FAprof) where: FA = aggregate capital, defined by the aggregate value of the fixed assets; T = aggregate turnover; 'prof' and 'loss' indicate the set for which the aggregation is made: profitable firms, unprofitable firms, respectively. T/FA is the capital productivity. All the above mentioned considerations for RK are also valid for RKPR. The examination of RKPR values allows us to make a few remarks. 1. RKPR values are generally less than one, meaning that the capital productivity is higher for the profitable companies. This is a predictable result. 2. There are also six exceptions more difficult to interpret (two for each ownership type): situations in which the capital productivity is not higher for the profitable firms. The case of Div18 should be mentioned, because is statistically significant: RKPR is 1.08 for FOR and 0.47 for domestic companies. The interpretation of these values is the following: the profitability and the capital productivity are independent variables (!) for the FOR companies (CPR = 1.08 doesn't significantly differ from 1), while for the domestic companies (STATE or PrivRO), they are dependent: the profitability is accompanied by a higher capital productivity. The other five exceptions are not statistically significant because of the small number of firms in the respective sectors. They should be interpreted on a case-by-case basis, which is not our objective. 3. Generally, the largest difference in capital productivity between the profitable and unprofitable firms appears in the FOR sector. Thus, at the level of the manufacturing industry, RKPR has the value of 0.46 for FOR, 0.51 for PrivRO and 0.72 for STATE. Also, the FOR sector locates the lowest values of RKPR at division level: 0,19 in Div 17; 0,08 in Div 28. These values indicate a high sub-utilization of the capital in the unprofitable area of the foreign sector. (Example: for Div28, the metal structures, metal products industry, except machines, equipment, installations, the capital aggregate productivity in the losing sector represents only 8% of the capital aggregate productivity in the profitable sector!) III.5 The share of the profit and losses in the turnover The analysis of profitability in §III.1 could be detailed, obtaining interesting general conclusions, but not information concerning the level of the profit in the profitable firms and the level of losses in the unprofitable ones. That's why we carried on a separate analysis for the profitable and unprofitable sectors (see columns 2-9 in Table A5). This approach starts from the observation that a discussion about competitiveness could be relevant only if, prior to it, the structurally noncompetitive firms (frequent enough) would be isolated. By structurally noncompetitive firms we understand one of the following two categories: 1) firms whose strategic mission is not the production and a profits transparently mention in the accountancy; 2) firms with negative or low profitability (as against the division they belong to), whose objective may be to make profits, but which can be viable on medium or long term only after a profound restructuring. It is, nevertheless, obvious that identifying structurally noncompetitive firm is a complex and also debatable problem which goes beyond the framework and possibilities of this study. The conjecture that we agree upon is the following: " the largest part of the structurally noncompetitive firms is to be found among the unprofitable firms". As a consequence, the analysis of competitiveness will be focused mainly on the analysis of the profitable firms sector. As a consequence, we shell focus the competitiveness analysis, mainly on the analysis of the profitmaking firms. Generally, this approach is not correct: the losses are not abnormal, they belong to the "rule of the game". The abnormal fact is the high frequency of the losses in the firms with at least 50 employees in Romanian '98 manufacturing: 33%. The conclusion is clear: as for the unprofitable firms sector, it cannot be ignored from the analysis due to its significant weight. Table III3. The appraisal of the profits and losses 0), 1) by prevailing ownership and NACE division Loss-making sector: Profit-making sector: Prevailing ownership ( profit 2) / losses 3) ) High losses divisions (more than 25% from turnover) Least losses divisions (less than 10% from turnover) Least profit divisions High profit divisions (more than 10% from turnover) STATE (4,56% / 17,66%) Div15, Div17, Div20, Div21, Div31, Div33, Div34 Div27, Div32, Div37 4) Div18, Div20, Div21, Div23, Div24, Div25, Div34 Div22, Div32