Firm-level determinants of exporting decisions in Greece

Transcription

1 Firm-level determinants of exporting decisions in Greece Vasiliki Skevi A dissertation submitted in partial fulfillment of the requirements for the degree of Master of Science in Applied Economics & Data Analysis School of Business Administration Department of Economics Master of Science in Applied Economics and Data Analysis September 2016

2 University of Patras, Department of Economics Author Vasiliki Skevi 20## All rights reserved

3 Three-member Dissertation Committee Research Supervisor: Dimitrios Tzelepis Assistant Professor Dissertation Committee Member: Nicholas Giannakopoulos Assistant Professor Dissertation Committee Member: Konstantinos Kounetas Assistant Professor The present dissertation entitled «Firm-level determinants of exporting decisions in Greece» was submitted by Vasiliki Skevi, Sid , in partial fulfillment of the requirements for the degree of Master of Science in «Applied Economics & Data Analysis» at the University of Patras and was approved by the Dissertation Committee Members.

4 Acknowledgments This dissertation encompasses the candidate s traits and competencies as they were formed by interaction with others and own self-struggle.i would like to thank Professors D. Tzelepis, N. Giannakopoulos and K. Kounetas for their unequivocal support during my studies. I would also like to extent my gratitude to all those who supported me along the way, each one with their unique manner. A special feeling of gratitude to my loving parents and many friends who have encouraged me throughout the process.

5 i Summary Using a large cross-section firm level database containing firms in greek economy for 2014, this dissertation estimates the impact of financial factors that affect exports decisions. Our sample is classified into three size-groups and each firm is identified with a unique industry code. An econometric study shows a picture that is consistent with related considerations. Our results highlight the importance of firms access to external financial resources on their decision to export. The analysis also suggests that financial constraints dipress selling abroad. Exporters display higher liquidity and leverage ratio than non-exporters. However, there is no evidence about ex-ante or ex-post firms financial health and how it could affect the results. Keywords: exports, financial constraints,extenal finance,firms size

6 Contents 1 Introduction 1 2 Literature review Determinants of the exporting decision: Related theoretical considerations Definition of variables Measuring financial constraints Data The dataset Descriptive Statistics Remarks Empirical model Modelling the decision to export Estimation methodology Results 27 7 Conclusions and Discussion 30 References 32 ii

7 List of Tables 4.1 Summary statistics firms group size Exports-firms size Summary statistics by group size Industry description Industry sector exporting decision Exports- liquidity Exports- leverage Exports- profit rate definition variables probit marginal effects : export participation iii

8 Chapter 1 Introduction It s a reality that Greece is an opened economy but considering its size that is not intensively integrated into the world market for goods, services or other factors of production. According to OECD survey in 2016 Greece does not play an important role in the global value chain. Export performance measured as the share of exports goods and services to export markets, deteriorated significantly in the last decade, particularly in service sector and much more than in Euro area on average. Export underperformance in Greece is estimated to account for a third of the decline in GDP between 2007 and Structural problems in product markets, barriers to exporting, access to finance and skills have outweighed the impact of large declines in wages on competitiveness. Developing exports further at a time when the domestic economy is expected to remain weak, can play a key role in reviving growth in Greece. Greek exports are dominated by oil (which is imported as crude, refined and then exported), shipping and tourism. The main goods exported are mineral fuels, food and live animals and manufactured goods. In services, tourism and transport are the most important exports (43%). Greece s main partners for both goods and services are the Euro Area and the other countries in the European Union. More specifically, in tourism, Germany, the United Kingdom and Russia represent 36% of Greek exports while the European Union as a whole represents 62%. The pattern of specialisation plays an important role in 1

9 2 constraining Greek export performance (Athanasoglou et al, 2010). Greek exports are concentrated in low-technology products which face strong competition from countries with lower labour costs, such as Bulgaria, China and Turkey. High and medium-high technology products, which have higher world market growth rates, are only around 20% of total exports compared to more than 70% in some OECD countries. More worrying, since 2007, almost all the increase in exports came from medium-low technology industries. Exports of the other industries have stagnated. The weak specialisation of Greek exports and dominance of low-tech goods partly explains the low integration in global value chains (GVCs). Global value chains allow firms and economies to specialise in the part of the production process they are best at, using intermediate goods and services from elsewhere without having to develop a whole industry. Being well integrated into a value chain enhances competitiveness by providing access to cheaper, more differentiated and better quality inputs (OECD, 2013b). Trade facilitating measures, such as fast and efficient port and custom procedures, are important for the smooth operation of value chains. According to the Bank of Greece, weak trade integration is consistent with the low Foreign Direct Investment flows into Greece; the small size of Greek enterprises; the small size of the manufacturing sector and lack of efficient infrastructure.the low-average firm size translates into a chronic lack of large firms. "This is because of those barriers of R and D and trade that are the main culprits that slow down firm growth. Countries that face higher trade costs provide fewer opportunities for businesses to become large. Multinational enterprises (MNEs) are the main actors involved in GVCs (OECD, 2013b), due to their international trade and investment links. It is often difficult for small firms, which are predominant in Greece, to enter international markets and value chains due to insufficient scale to support; the costs of adequate R and D; the training of personnel and the fulfilment of strict requirements in terms of product standards and quality (OECD, 2013b). Upgrading a small firm s position in the value chain requires the adoption of a wider and more complex set of tasks. Small firms can also participate indirectly in GVCs

10 3 as providers of bigger exporting firms (G20 report). OOSA survey prove that the highest share of exports is from large (250 employees and more) enterprises. Nonetheless, the share of Greek SMEs (defined as firms with 0-50 employees) in exports is higher than the OECD average (OECD, 2014). SMEs represent 99%of firms in Greece and employ 80% of the labour force. While the degree of export orientation in Greek firms has marginally increased between 2008 and 2013 (from 8% in 2008 to 9% of turnover in 2013), the average level of exports by SMEs has declined over the same period. Survey results show that larger SMEs were more resilient during the crisis, they turned more to growth strategies and face limited liquidity problems. But smaller SMEs struggle to survive and 56 % continue to face severe liquidity problems (NBG, 2014). Manufacturing SMEs have the greatest export intensity. Almost 16% of SMEs are potential exporters, i.e. they do not currently export, but intend to export if certain obstacles were removed. The main obstacles are i) institutional inflexibilities such as excessive bureaucracy and inefficiencies in customs procedures, albeit the latter has been reduced notably by the introduction of electronic procedures; ii) poor access to special financing for exports (such as trade credit); and iii) shortcomings in distribution networks and transport infrastructure. Policies that create a business environment where firms can easily enter (and exit) the market and young high-performing firms can thrive and grow should be assessed with particular care. Such policies include both sticks (i.e. regulations which only affect firms above a certain size) and carrots (i.e. support mechanisms for which only smaller firms are eligible). In this context, resources will not be trapped in small and inefficient firms. Further liberalizing product markets and better bankruptcy procedures will help in this respect. Unit labor cost tends to decline since the beginning of crisis and has restored cost competitiveness but the response of exports has been sluggish due to severe liquidity constraints of exporters. Furthermore prices did not adjust as fast. Price competitiveness improved by 5% in Greece between 2009 and Non cost competitiveness is also weak in Greece, as Greek goods exports are

11 4 concentrated in low technology products. Researchers should explain firm-level financial conditions that create difficulties to exporting activities in Greece. Most studies ignore heterogeneity. As a consequence there are problems in model implications either on planning or on evaluation of economic policy. This study proposes a parsimonious theoretical model of financial constraints, and determinants of exporting, where financial factors may influence the exporting decision. This study adds the element of firm heterogeneity, the financial dimension. In particular, we focus on whether firms financial health and their ability to enter export markets are linked. In so doing we build on an extensive literature that has focused on the effects of capital market imperfections on firms activities. Within this literature, a general conclusion has been that financial constraints affect firm investment. Our analysis is based on Greek firms from all Greek industries in The model suggests that financial factors and other firm s characteristic are reflected in or driven by six endogenous variables: Size, market share, internal funds (liquidity), and the costs of external finance (leverage) and profitability. Empirical counterparts to those variables that are widely employed are firms cash ratios, sales (size or market share), the dept ratios such as long term bank liabilities and short-term bank liabilities relative to firms leverage ( for costs of external finance). Focusing on Greek firms can be motivated by the consideration that Greece geographical position plays an important role not only for Europe but for the international trade. The results suggest the difference between exporters and no exporters to their financial performance. The rest of the paper is organized as follows. I next present an overview of the literature on financial constraints, firm s characteristics that affect the export behavior. Section 3 presents the required variables to test the econometric model. Section 4 presents data and summary statistics for this essay. Section 5 discusses the shortcomings of usual strategies employed to measure financial constraints, firms financial health and illustrates the methodology adopted here. We then look more in detail at the role played by financial variables in shaping the decision to

12 export: these results are discussed in Section 6. Section 7 concludes and draws some policy implications. 5

13 Chapter 2 Literature review This study is related to earlier research in at least two aspects. First, it adds to the empirical evidence on international trade, the present study motivates effects of such constraints on firm-level exporting propensity. Second, it contributes to the literature on Greek firms particularity. In a country where 85 % of private employment is concentrated in SMEs and more than 50 % are micro enterprises (0-9 employees), a prolonged recession, which is exacerbated by austerity measures and the delay of absolutely vital structural reforms, has affected SMEs profoundly and disproportionately more than large enterprises. In that regard, the study illustrates that financial constraints affect exports along with their total business performance during the crisis. Third, it relates to the empirical literature on the effects of financial variables on exporting, utilizing a special framework for causal analysis. 6

14 2.1 Determinants of the exporting decision: Determinants of the exporting decision: The significance of export decision is evidenced by the extremely large number of research studies that have been conducted on the specific area in international literature. Exporting decision of firms differ because of financial constraints, and other firms characteristics such as their industry, their size or the market share. Why?

15 2.2 Related theoretical considerations Related theoretical considerations A starting point is the stylized fact that exporting tends to be concentrated in the larger production units in an industry. Size therefore could be both cause and effect of export performance. The estimated empirical model from Wagner (1995) in 7000 German firms proved that the probability that a firm is an exporter increases with firm size. Wagner classified firms into three size classes (small firms with up to 10 million euro of sales, medium firms with 11 to 50 million euro of sales, and large firms with more than 50 million euro). The importance of size as a determinant of exports follows from economies of scale in production, a more fully utilization of (specialized) executives, the opportunity to raise financing at a lower cost, benefits form bulk purchasing, own marketing department plus own sales force, and a higher capacity for taking risks (e.g., development of new products) due to internal diversification. Obviously, there are limits to the advantages of size, because coordination costs mount as the scale of operations increases, and at some point further expansion ceases to be profitable. Furthermore domestic share has a positive influence on the export performance of a firm. It s a comprehensive theoretical model for the export decisions of a firm that discriminates between exporters and non-exporters. Too, from an alternative interpretation of a high domestic market share as a constraint imposed on a firm by a domestic market too small to expand output further, so that firms with a high market share are obliged to export if they want to grow. Recent studies have shown that smaller firms tend to be less productive (OECD, 2014). Firms with high productivity tend to engage in exporting. Fixed costs associated with the penetration of foreign markets discouraged them from exporting. Firms manage to recover those costs they could become exporters (Melitz, 2003; Melitz and Ottaviano, 2008). Such costs are related with the required information, commercial contracts or hiring multilingual staff or adapting products to be sold abroad (Arnold, 2015). A large part of the research on export and credit constraints focuses at the firm level. Exporters own

16 2.2 Related theoretical considerations 9 more liquidity than domestic firms. Greenaway et al. (2007) study the causal relationship between export and financial health using a panel of UK manufactures. He found that exporters exhibit better financial health than non-exporters. The element of firm heterogeneity in this paper is the financial dimension. He suggests that export promotion policies could be helpful, reducing the level of financial constraints faced by firms, and indirectly enhancing their investment spending and productivity. The latter effect could be particularly relevant for small and medium sized enterprises (SMEs), whose investment is often constrained by lack of finance. The concern around export and credit is more and more increasing, especially because of the policy implications of results involving the banking and production sectors simultaneously. Internal resources are an important factor for the internationalization of firms and in particular that entry decisions are determined by the level of cash stock for those firms identified as credit-constrained. Considering export s entry costs as an investment, it appears that internal liquidity may affect entry choice. (Forlani 2010). In this paper Forlani focus on testing the constrained firms by measuring their liquidity and leverage ratio. Bermann and Hericourt (2010) find evidence that credit access is an important factor in determining entry into the export market for firms in developing countries. They also measured liquidity and leverage ratio. The role of finance on trade is mainly concentrated at the time of entry. The present work points out the importance of considering heterogeneity in terms of access to external finance. Bellone et all claim that financial constraints represent an additional source of heterogeneity that can help to account for different in export behavior across firms and their analysis is based on the theoretical background of firm heterogeneity and the relevance of sunk entry costs into export markets. They find strong evidence that less credit-constrained firms self-select into export markets, or from a complementary point of view, that external funds are an important determinant of firm export status. Access to external financial resources is a significant but not crucial determinant of the probability to start exporting, but they find no evidence of a

17 2.2 Related theoretical considerations 10 positive relationship between financial health and the share of production sold abroad. The main reason why financial constraints may have an important effect on exports is related to the existence of fixed costs of exports. Wagner (2011) reports that the probability of exporting and the export volume increase with age which is consistent with the hypothesis that firms need to accumulate sufficient collateral before they can borrow enough funds to profitably export. Minetti and Zhu (2011) analyze a cross sectional survey of Italian manufactures and find that credit rationed firms are less likely to export. The probability that a firm is credit rationed is related to some attributes like liquidity and external finance. Limited access to liquidity appears to impact both the probability that a firm exports and firm-level exports. Finally,when we dig deeper into the relation between financial factors and the decision to start exporting, we find that better access to financial markets increases the probability of firm internationalization. Generally, there are not many empirical studies for Greek exporting performance in firm-level. Nowdays, researchers are more and more occupied with this subject but they should overcome huge problems in order to take valid results.

18 Chapter 3 Definition of variables Exports declare the firms participation in foreign markets, by a dummy variable that takes on the value 1 if a firm is an exporter and 0 otherwise, is included in the empirical model as the dependent variable. Export strategy is vital to a country s economic survival and is considered to be a core building block for achieving economic development and growth (Kedia and Chokar, 1986, 33; Soontiens, 2002, 710). In countries with small population there is a further incentive to export in that the domestic market may not provide economies of scale and selling abroad may help (Ramaseshan and Soutar, 1996, 53). Firm size is measured as the logarithm of total assets of an establishment during the given year. Because of the lack of information about the number of employees in each establishment, it couldn t be possible to measure firm in this way. Studies and literature have shown that the impact of firm size is positive showing that the probability that a firm is an exporter increases with firm size and most of the exports are from the top size groups of firms. Wagner s paper has explored the relationships between exports, firm size and firm dynamics. An alternative approximation is the firms classification into three size groups according to their sales. The Definition of the Euro- 11

19 12 pean Union according to European Commission describe small firms with less than 10 million euro sales, medium firms with 10 to 50 million euros, and large firms with more than 50million euros. Furthermore according to Berman and Hericourt firm size coefficient is significantly high in more financially dependent sectors. Market share is the percentage of market s total sales that is earned by a particular company over a specified time period. Market share is calculated by taking the company s sales over the period and dividing it by the total sales of the market over the same period. Particularly a firm that has increasing market share will be more likely to export than another firms with decreasing market shares. One of the most frequently used tools of financial ratio analysis is profitability ratios which are used to determine the company s bottom line and its return to its investors. Every firm is most concerned with its profit. Profitability ratios show a company s overall efficiency and performance. Profitability of a firm is computed as a rate of return defined as profit before tax and interest divided by total sales or by total assets. Wagner s and Fryges paper documents that the positive profitability differential of exporters compared to non-exporters is statistically significant. However, previous studies have shown that exporting activity is not systematically associated to higher firm s profitability. This is shown both by means of non-parametric methods and, with an approach that is more standard within the empirical trade literature, by regression techniques that try to identify an export premium (Marco Grazzi,2010). In this case we test this hypothesis in order to prove the validity of this fact for the Greek firms.

20 3.1 Measuring financial constraints 13 Liquidity is calculated by taking the company s current assets over the period and dividing it by the current liabilities over this period. Only firms that own sufficient liquidity or access to external finance will be able to enter. In this case, access to finance may play a role. The more the short term liquidity the firm has the more likely it will be able to enter the export market. The coefficients on the interaction term between firms liquidity are positive and significant in most of the previous specifications. Financial leverage ratio takes as a numerator the total assets and as denominator the total equity or respectively total dept and total assets. On average credit rationed firms have a higher leverage ratio than non-rationed firms. The probability of exporting is higher for firms with a higher leverage ratio. The effect of leverage could suggest that highly indebted firms use exports to shift the risk associated with their high leverage to creditors. This could be also happened because of the fact that these firms had to draw down liquidity and increase leverage to pay the sunk costs necessary to enter export markets. Dataset also includes information about long term liabilities, bank liabilities, EBIT for each establishment that will help to explain their export performance and the exporting decision. 3.1 Measuring financial constraints The items used to operationalize this construct were developed on the basis of existing literature. In the essay, we experiment different measures of financial constraints. The first two are the liquidity ratio and the leverage ratio as employed by Greenaway et al. (2007) we find two main shortcomings in these measures. First, they only capture one dimension of access to financial markets: a firm may

21 3.1 Measuring financial constraints 14 be liquid but nonetheless present a bad financial situation; on the other hand, strong fundamentals may compensate for a temporary shortage of liquid assets. Second, both ratios may suffer from some endogeneity. In other words, there are no clear-cut theoretical priors on the relation between either liquidity or leverage and financial constraints. While liquidity is generally regarded a sign of financial health, firms may also be forced to withhold cash by the fact that they are unable to access external funds. In fact, a recent theoretical contribution by Almeida et al. (2004) shows that financially constrained firms tend to hoard cash, so that liquidity would be associated with financial constraints, not lack thereof. In a similar vein, a high leverage, while signaling potential dangers, suggests also that the firm has enjoyed, at least in the recent past, wide access to external financial funds. Hence, one could argue that highly leveraged firms are not financially constrained. Specifically after following the above literature this paper explains liquidity as the current ratio taking into account the current assets and current liabilities. While on the other side, leverage ratio, measured by total dept over total assets.

22 Chapter 4 Data 4.1 The dataset The study assess the role of financial factors on exporting behavior at firm data. The data employed in the analysis are provided by "hellastat" for imentor database. It s a database that collects data for more than Greek firms. It provides us many financial variables and ratios for each firm. The data cover a sample of small, medium-size and large enterprises in different sectors. It covers food industry, industry goods, tourism, consumer and business services, energy and utilities, pharmaceutical goods and technological good and services. The data from the surveys and especially exports variable are not time-variant, so the empirical analysis uses cross-section techniques. In this essay we used microlevel data for enterprises in 2014 and exports will be stable over this period. The key information about short-and long-term debts, credit, assets, equity. Finally, firms are classified according to a one-digit industrial classification; sector codes and the descriptive statistics on the sector level. The survey asks if a firm exported to one of nine regions of the world during the period under consideration. So this study analyzes only the probability of the decision of a firm to export. There are no elements of exports to sales ratio in this database so it s difficult to use another method in order to come to safe conclusions.in Greece We haven t sufficient empirical evidence from the econometric investigations on the causes 15

23 4.2 Descriptive Statistics 16 and consequences of exports based on comprehensive micro level data sets until now. A reason for this "state of the art" surely is that suitable data sets are not available for public use, and not collected by researchers due to budget and time constraints. However, the ELSTAT collects firm level data on a large number of topics, including exports. For obvious reasons, these data are not published at the firm level. Because of many complicated and expensive procedures we couldn t collect these data elements. Firms that did not have complete records on exports, assets, equity and the relevant financial variables were also dropped. As a consequence the beginning sample was about firms but only had the acquired records to take valid results. 4.2 Descriptive Statistics Table 4.1. Summary statistics Variable Mean Std. Dev. N ROA MS liquidity leverage _Iind_ _Iind_ _Iind_ _Iind_ _Iind_ _Iind_ _Iind_ _Iind_ _Iind_ small medium large Source:Hellastat-imentor Table 4.1 reports means and standard deviations of key variables identified in the literature as determinants of export participation, as well as of two variables which capture firms financial health: the liquidity ratio and leverage ratio. The

24 4.2 Descriptive Statistics 17 former is defined as the firm s current assets less current liabilities over total assets, while the latter is defined as the firm s ratio of debt over total assets. We also report mean and standar deviation of a measure of firms profitability. Return on assets(roa) points out the profit rate of each firm. Table 4.2. firms group size Small Medium Large Total 92% 6.2% 1.8% 100% Source:Hellastat-imentor Table 4.3. Exports-firms size Exports Small medium Large 0 89% 60% 60% 1 11% 40% 40% Total 100% 100% 100% % Source:Hellastat-imentor Table 4.3 reports the decision of firms to export according to their size. The classification into three groups small, medium and large depends on their turnover percentage. Small firms operate with less than 10 million euro turnover. Greece has a great majority of small firms and most of them are microbusiness. Most microenterprises are family businesses employing one or two persons(iobe). The lack of data allows us to include only companies from the sample of establishments. Small business possesses the 92% of the remaining sample from which only 11% decided to export. Medium size firms have more than 10 million euro turnover and less than 50 million. They are approximately the 6,5% of the sample. About 40% of medium size firms are positive to export participation. Finally a firm that is part of larger group has more than 50 million turnover which are only 2% in the sample of businesses. In this sample 40% of the larger group sold part of their products abroad.

25 4.2 Descriptive Statistics 18 Table 4.4. Summary statistics by group size small mean sd liquidity leverage ROA MS medium liquidity leverage ROA MS large liquidity leverage ROA MS Source:Hellastat-imentor Table 4.4 presents the key variables in our sample for each group size. We took into consideration the mean, standard deviation, minimum and maximum. Concerning liquidity, it is observed that the larger the firm, the larger will be the liquidity ratio. In other words the proportion of the variance in the change of liquidity ratio is expected to be higher for the sample of large firms than for medium firms and larger for medium firms than for small firms. Without cash coming in the door, they can quickly get into trouble with their creditors. Possibly larger firms due to the fact that they have to pay more to creditors need to have a large amount of cash in order to take care their liabilities. Firms with higher liquidity ratio are less financial constrained. On the other hand leverage ratio dimension allow us to study the relationship between the businesses group size and the degree of dependence on external funding. As shown in table 4.4 small firms have the lowest leverage ratio. The larger the firm the larger will be the dept in comparison with the total assets. The appropriate amount of leverage for each establishment will be based on their risk appetite. An aggressive trader may utilize effective large leverage amounts than someone more conservative trader.larger firms used to be exposed at high risks, enjoying supreme efficiencies. Concerning ROA index there are various result of the effect between firm size and profitability.

26 4.2 Descriptive Statistics 19 Studies have shown positive influences between firm size and profitability. The larger the firm, the better its access to the resources the more likely it is to increased profitability.larger companies tend to sell larger volumes of products relative to what smaller companies can sell (obviously, in absolute terms as well). They are also able to negotiate volume discounts as well as payment terms. This allows them to finance sales through their supply chain. Since performance of large size companies are better than small size companies enjoy a higher level of profits.mean profitability for large businesses of this sample is about five times more than small business profits. As far as the firm market share is concerned there are grounds for believing that a significant, but not very strong, relationship exists between size and concentration. Concentration (and firm size) leads to "economies of scale" which smaller firms in the same market cannot gain access to. Market share is directly related to the size of enterprises. Furthermore it is now widely recognized that one of the main determinants of business profitability is market share. Under most circumstances, enterprises that have achieved a high share of the markets they serve, are considerably more profitable than their smaller-share rivals. Hence the claim that the fact that the largest firms tend to be most efficient and hence more concentrated is real.particularly the market share for large dummy variable in our sample is slightly higher than that of medium dummy variable. For small businesses the market share is equal to 0.

27 4.2 Descriptive Statistics Remarks Table 4.5. Industry description Industry code Industry description firms percentage in sample 1 Business Services 11% 2 Consumer goods 21.6% 3 Consumer services 0.4% 4 Energy sector 3% 5 Food sector 10.5% 6 Industry goods 30% 7 Pharmaceutical products-health care services 6% 8 Technological goods 3.5% 9 Tourism services 15% Total 100% Percentages of observations of the sample for each industry, Source:Hellastat-imentor Table 4.6. Industry sector exporting decision Industry code Exports= 1 3% 27% 2% 1.5% 16% 41% 5% 4%% 1% Percentages of exporters of the sample for each idustry, Source:Hellastat-imentor The sample consists of businesses of which only 4794 of them export. Number 1 industry code refers to business services firms. The code 2 refers to firms that occupy with consumer goods, code 3 includes consumer services firms, code 4 refers to energy sector, code 5 is about food sector, code 6 includes firms that produce industry goods, code 7 has to do with pharmaceutical products or health care services, code 8 is related with companies that produce technological goods and code 9 refers to tourism services.specifically, 3% of firms decided to export in 2014 from business services sector. About 27% of firms that produce consumer good export. In food sector about 16% are exported orientated. Also 41% of firms export industry goods, 5% export pharmaceutical products and health services, 4% export technological nature products and services. Only 1% from companies from tourism sector have been exported at least one time

28 4.2 Descriptive Statistics 21 since Furthermore about 86 consumer services firms and around 57 from energy nature businesses have positive export behavior. Other 140 businesses have not identified which sector they belong to but they are orientated to exports.the predominant sub-sector of economic activity among exporters is production of industry goods followed by consumer goods and food products. Table 4.5 also shows the distribution of the foreign industry sectors of the Greek firms in our overall sample. Foreign businesses are mostly located in industry goods (30% of total sample) and consumer goods about 21% of our sample.

29 4.2 Descriptive Statistics 22 Table 4.7. Exports- liquidity Exports Mean liquidity freq Source:Hellastat-imentor Table 4.8. Exports- leverage Exports Mean leverage freq Source:Hellastat-imentor Table 4.7 shows that exporters have higher liquidity than non- exporters in our sample of establishments. Firms that are able to export have sufficient liquidity to finance the fixed cost in order to access the foreign markets.it is obserevd that non-exporters deal with negative liquidity ratio. As a consequence non exporters are liquidity constrained. Hence increasing financial constraints leads firms not to export in a foreign market. Table 4.8 It is observed that export market participation assumes that firms have a higher leverage ratio. Maybe these firms increase leverage to pay the sunk costs necessary to enter export markets. So from this ratio it is expected a positive relationship between export participation and leverage. Measuring profitability, is one way to identify each firm s financial health. Table 4.9 shows that exporters seem to have better financial performance than non- exporters. There is a huge difference between the mean average profitability of exporters and non-exporters. Table 4.9. Exports- profit rate Exports Mean profit rate freq Source:Hellastat-imentor

30 Chapter 5 Empirical model 5.1 Modelling the decision to export Firm export behavior must ultimately be conceived as a series of decisions regarding both participation to export markets and the firm s commitment to international trade. These decisions can be modelled as the outcome of a variety of factors. Heterogeneity of firm profitability levels is the utmost explanation for the observed differences in export behavior across firms because of the observed differences in productivity levels. Nonetheless, there was lack of data for enterprises so as to calculate productivity levels. That s the reason of using profit rates. Yet this may not explain entirely the decision. The firm s ability to access external financial resources may well constitute another important part of the story. Another determinant could be liquidity ratio or firms size. In this Section, we investigate the factors driving the decision to export in a standard binary choice framework. Taking stocks of our previous findings, we expect financial constraints to be an important driver of export behavior by firms, controlling for other relevant factors such as profitability, liquidity, firm size and market share etc. 23

31 5.2 Estimation methodology Estimation methodology For the econometric analysis, we have employed the empirical model that will examine the factors influencing the export performance of all Greek firms. The export behavior will be reflected by one decision; the probability of exporting (export propensity).on the basis of previous empirical findings and theoretical argumentations regarding the determinants of export participation, a meaningful and informed set of explanatory variables is included among the available economic and financial variables that also capture the financial constraints. The econometric analysis is therefore framed to show that the decision can potentially be dependent. The specified model evade selection prejudices which can occur if focus on the probability of exporting. Thus, the methodology adopted can give some credibility to the results obtained in this study. exp i = α + β 1 ROA i + β 2 liquidity i + β 3 leverage i + β 4 MS i + β 5 small i + β 6 medium i + β 7 large i + β 8 ind4 i + β 9 ind5 i + β 10 ind6 i + β 11 ind9 i + u i Where EXP=1 if firm exports EXP=0 otherwise And the error term (ei) is normally distributed with mean zero and variance one and the subscripts, i refer to the firms. For this study, the variables included in the estimation equation were selected from the database depending on the response rate to the above equation concerned. Due to the high rate of missing data on some key variables that we include in the regression equation, only observations from the sample have all the required data so as to run the regression.. The fixed costs of exporting are assumed to be equal across firms. A detailed definition of the variables is given in

32 5.2 Estimation methodology 25 Table 5.1. definition variables Variable Definition Exp discrete variable equal to 0 or1 ROA(profitability) ProfitBeforeTax/Total Assets liquidity (Currentassets- currentliabilities)/ Total Assets leverage Debt/Total assets MS company s sales/market s sales small dummy variable for firms with less than 10 mil euro of sales medium dummy variable for companies with sales of more than 10 million and less than 50 million euro large dummy variable for companies with sales of more than 50 million euro _Iind_4 dummy variable/energy sector _Iind_5 dummy variable/food sector _Iind_6 dummy variable/industry goods _Iind_9 dummy variable/tourism Source:Hellastat-imentor Table 5.1: We have little evidence about changes in export behavior because of return on assets. Liquidity and leverage express constraints. Liquidity effects is expected to be positive according to previous studies (David Greenway,2007). Furthermore leverage ratio has been examined by different aspects so that results are controversial. On the one hand less leveraged firms are generally more likely to export(greenway,2007) but on the other hand more levered firms sell more abroad. (Minetti and Zhu,2011) In other words, the proportion of the variance in the change of exporting probability that is explained by the variance in the growth of total sales is expected to be higher as the firm s size is increasing(wagner,1995). Market share that declares in one way firm s size is not often met in literature but it has been proved the positive effects on export behavior by Wagner. That fact in combination with the classification into group size according to European Commission shows that we expect similar impacts on exporting decision. Nevertheless because of Greek business distinctiveness, we have to examine the impact of the main and more profitable sectors as energy, food, industry goods and tourism. Our estimates of equation are reported in table 5.2 using a number of estimators. Specifically they represent coefficients on expi where i indexes the firm in 2014, in a cross- sectional regression.

33 5.2 Estimation methodology 26 The coefficients of interest are b1, which capture the effect of profitability in year 2014 on exports participation probability, b2 shows the effect of liquidity, b3 reports to leverage impact. The other coefficients refer to group size and industry dummy effects. Table 5.2. probit marginal effects : export participation Variable Coefficient (Std. Err.) ROA 0.198* (0.010) liquidity 0.550* (0.13) leverage 1.153* (0.428) MS ** (24.061) small * (0.036) medium 0.106*** (0.037) o.large (0.000) _Iind_ *** (0.006) _Iind_ *** (0.007) _Iind_ *** (0.004) _Iind_ *** (0.003) Intercept *** (0.154) Note: Marginal effects computed at means, Significance levels:p < 0.05, P < 0.01, P < Source:Hellastat-imentor

34 Chapter 6 Results In this study, the econometric investigation has been to regress the probability to export to examine the intensive margin of trade, on opportune measures of firms financial constraints. Credit rationing at the firm level is approximated using either financial sheet variables or specific information directly gathered through our database. In order to deal with the endogeneity issues, this paper usually utilize instrumental variable method. The econometric investigation provided mixed evidence on the role of financial constraints on firm trade performance. The marginal effects (average marginal effect) of the above equation are directly reported in Table 5.2; then we can interpret the coefficients as elasticities(i.e., variation in the probability of export participation due to variation in the variable of interest). Exporters always display higher liquidity and leverage compared to those that did not export. All reported coefficients are significant. Therefore better financial health leads to exporting decisions. Moreover given the sunk costs need to be met when entering foreign markets for the first time, being an exporter also provides a signal that the firm is sufficiently liquid to generate enough profits in foreign markets to recover these. This increase the likelihood that the firm will be able to service its external dept. Probably the positive relationship observed between firms financial health and export status is therefore driven by the fact that exporting improves ex-post financial health. Maybe this is a two-way relationship. 27

35 28 The impact of liquidity on the probability to export should increase by increasing assets and decreasing liabilities. A positive sign will mean that the exporting probability increases. Then if the cash stock increases 10%, the export probability grows by 5,5%. Thus firms need resources to cover fixed entry costs if they want to export; because those firms that experience difficulty securing financing from external investors, they rely on internal financing. Secondly, the leverage ratio has a positive impact on export probability: more levered firms can mean a larger pool of potential investors. The effect of the external credit is positive and significant. The interpretation is quite straightforward: as external finance increases, the entry probability increases. In case that total foreign funds increases 10%, the export participation grows by 11,5%.Also firms in order to increase their extensive margin, need resources that can be used to cover past debts if their financial situation is potentially critical; this may explain the significant positive coefficient. Bank or other external financing has allowed business to increase their size and, thereby, to reap whatever economies of scale have long been available to larger entities by easing restrictions on their abilities to acquire other financial institutions and to operate over broader geographic areas. Furthermore the estimated coefficient of ROA index shows that profitability can be interpreted as a causal effect of changing exporting decisions. Calculating this effect reveals that an increase of 10% of ROA index is significantly increasing the probability of the establishment to export by 1,9%. In other words, larger net profits will encourage firms to export in a foreign market.also this has been shown by descriptive statistics of the previous chapter that exporters seem to be much more profitable than non-exporters.as displayed on table 5.2, the effect of size classes on the probability of exporting appears to be significantly across firms of different size. Small firms have a negative impact on export probability in common with early findings. Maybe this happens because small firms deal with more financial constraints than for big ones. An increase of small class about 10% will bring 0.8% decrease of export participation. On the other hand the probability that a

36 29 medium size firm is an exporter increases significantly at 1% when medium class increases by 10%. Market share is another instrumental variable that express a different measure of firms size. It has been proven that positive and significant effect on export participation. An increase of business market share will increase the probability to sell abroad. For industry dummies all reported coefficients are strongly significant. Firms that belongs to the Energy and the tourism services have negative effect for firms exporting decision. However, food industry and industry goods have positive effects on exporting probability. A firm that belongs to the last two sectors is more likely to be an exporter than those firms that belong to energy or tourism sector.

37 Chapter 7 Conclusions and Discussion In this essay we used a detailed unique set of micro data for some Greek firms from the establishment level of Hellastat database. We have introduced a financial dimension to analyze the role played by financial variables in determining firms decision to export. We found that less financial constrained firms are more likely to export. On this basis exporters exhibit better financial health than non-exporters. Results may suggest that firms in the sample are quite small and find an advantage in using external funds to finance their investment exporters are relatively more leveraged than other firms. The existence of sunk cost associated with exports forces firms to use external funds for part of their financing. Similarly, we found that exporters deplete their financial stability, increasing debts to overcome these costs. In addition liquidity efficiency give another incentive to firms for selling abroad. Applying cross section analysis we find that the coefficient of profitability has the theoretically expected positive sign for exporting decisions. Also data shows that the probability to export tends to increase with firm size and market share. From the empirical model estimated a picture emerges that is consistent with theoretical considerations: the impact of firm size and market share are positively related to the export performance of firms. Furthermore results presented here shed some light on important industry sectors for Greece. However, tourism and fossil fuels until now are generally profitable our study 30

38 31 proved that in 2014 those firms that belong to these sectors are discouraged from exporting decision. Maybe it is connected with the fact that many problems came from greek goverment policy and other competitive disadvantages for greek firms. These findings suggest that export promotion policies could be helpful forcing those firms to ameliorate their services. It s of great importance that small and medium sized enterprises (SMEs) are the backbone of the European economy and we look to our small companies to get us out of the crisis, yet large businesses still contribute disproportionately more to greek economy. Unfortunately, this reaffirms the message that we have been voicing for years: cut down on the regulatory and administrative burden for SMEs, improve access to and cost of finance and allow small businesses to prosper, innovate and grow," Caro said. More could be done to promote exports and help SMEs reach international markets. Enterprise Greece, the recently created export and investment promotion agency deploys annually an Action plan for exports promotion and is responsible for promotional activities, branding, organization of conferences and development of domestic and international networking. The new export promotion action plan aims at addressing several issues that emerge from difficulties in exporting procedures. In Greece the emphasis should also be on cutting trade costs and improving access to international markets. Nations do not produce, do not trade, do not compete; it is firms that produce trade and compete. This simple truth makes it clear that understanding the firm-level facts is essential to improve policy making. Firmlevel analysis should complement the policy making toolbox in Europe. Then we should expect that entry into foreign markets will become more attractive for the majority of business.