80 minute read

INDUSTRY IN SLOVAKIA

Next Article
SLOVAK CUSTOMERS

SLOVAK CUSTOMERS

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 64(1): 123−133, 2022 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2022.64.1.11

SPECIFICS OF FAMILY BUSINESSES IN THE WOODWORKING AND FURNITURE INDUSTRY IN SLOVAKIA

Advertisement

Anna Kocianová – Mariana Sedliačiková – Jarmila Schmidtová – Mária Moresová

ABSTRACT

A survey intothe field of the family business in woodworking and furniture enterprises in Slovakia has not been carried out yet. The presented paper tries to eliminate these shortcomings. Its goal is to map the situation of family woodworking and furniture enterprises in Slovakia as a hitherto unexplored segment to capture a view of their current position together with the main internal and external determinants hindering their development in practice. Based on the results obtained by conducting and evaluating the questionnaire, it can be stated that more than half of enterprises consider a lack of qualified workforce and increasing intensity of competition to hinder their development along with specific problems arising from the nature of family business, which is the underestimation of the issue of succession. A contribution at the level of theory and practice is assumed in the paper as well. The main contribution is characterizing of a family business as the legal definition in the legislative conditions in the Slovak Republic is absent. The contribution also refers to the future direction of the development of the Slovak woodworking and furniture family enterprises. Key words: family business, furniture industry, specifics, woodworking industry.

INTRODUCTION

Family business (FB) faces, not only in Slovakia, insufficient legislative support (MORESOVÁ et al. 2020). The statutory definition of a family business in Slovakia is absent, and therefore, based on its own survey from 2020, the Slovak Business Agency (hereinafter SBA) defined it for domestic conditions as follows: “Family business means a group of natural persons who have blood ties or court decisions or legally recognized ties with the same effect (adoption, marriage) or have personal, mutual ties and are interested in creating interdependence, coherence and dependency through these common ties (partner) and meet at least one of the following conditions in relation to the business: • one or more members own more than 50% of the shares, votes, or stocks in the company, • one or more members own such a number of shares, votes or stocks that they can enforce their will against other co-owners (hereinafter referred to as “has influence”),

123

• one or more members perform control functions in the company and have influence, • one or more members perform managerial functions in the company and have influence“.

According to the above definition, the share of FBs in Slovakia is estimated at 60 to 80% in all sectors. FBs in Slovakia produce 30 to 40% of GDP and provide 40% of employment (PERÁČEK et al. 2020). The data presented by the European National Association representing FBs in the EU (EFB) are even more favourable. GDP indicator is estimated at 40% and the employment rate at up to 60% (SHARMA 2004).

Family business has a long tradition. It dates back to 587 in Japan, where the first family business was established (KRISTIE 2002). Since then, the importance of family business has been growing and nowadays, family businesses make up more than two-thirds of all businesses worldwide. Slovakia, Cyprus and Estonia reach the highest share in the European Union (in all more than 90 %) (MURA 2019).

Family businesses have many advantages as well as disadvantages, which can be considered the specifics and differences compared to non-family businesses. The primary ones include the following: the owner usually also acts as a managing director, the family's income depends on the success of the FB, most FBs have a family name included in the denomination, the FB owners expect a generation change. Disadvantages can be considered: strict adherence to quality and reputation at the expense of other indicators (gain, profitability, etc.), conflicts of family members, the selection of a manager from among family members and the above-mentioned advantages in reverse principle (i.e. family collapse in case of failure, lacks in management in case of insufficient qualification of family members and others) (HENNART et al. 2019, ARRAGLE et al. 2016, MARINOVA and MARINOV 2017, SCHOLES et al. 2015, DE MASSIS et al. 2016, VERBEKE and FOROOTAN 2012).

The woodworking and furniture industry, together with the pulp and paper industry, form a uniform sector of the wood-processing industry (WPI) in Slovakia, which has a long tradition and has hitherto unused potential (MORESOVÁ et al. 2019, HALAJ et al. 2018). The rich domesticbaseofraw material, processing capacities (especially in the case of coniferous log processing), a stable share of the industry in the field of industrial diversification (approximately 2.5% over a decade) are some of the many predispositions. Long-term problems of the industry are the lack of support and development strategy, lack of targeted modernization of technology and machinery, low attractiveness of the industry from the point of view of investors and entrepreneurs themselves, insufficient product finalization, WPI enterprises act as subcontractors of semi-finished products for foreign companies, slow growth of competitiveness (MUSOVA et al. 2021, HAJDÚCHOVÁ and HLAVÁČKOVÁ 2019). At present, more than 5,300 business entities are operating in this segment, which follows the tradition of WPI in Slovakia, compared to the European average, which is more than 397 thousand business units (19.6% of all businesses) (KRIŠŤÁKOVÁ et al. 2021, ANTOV et al. 2020). The level of forest cover in Slovakia, which reaches 41.2% and has been growing exponentially over the years, also contributes to their prosperity (ŠEBEŇ et al. 2018).

The aim of this paper is to map the situation of family woodworking and furniture businesses in Slovakia, as a hitherto unexplored segment, to define their current position together with the main internal and external determinants hindering their development in practice.

METHODOLOGY

Achievement of the set goal presupposes scientific work at the level of both theoretical and practical in the form of obtaining primary data. The first step was to define the FB for 124

the needs of this paper, based on the definition proposed by the SBA (2020), methods of analysis, description, comparison, analogy, summarization and synthesis. Mapping of the situation of woodworking and furniture family businesses in practice presupposes the acquisition of primary data, for which a questioning method was used in the survey in the form of an electronic questionnaire (SCHEER 2007, RIMARČÍK 2007). The questionnaire contained a total of 29 questions (closed and semi-closed with the answer type “other”).

Inner consistency of a questionnaire was evaluated by the Cronbach alpha coefficient (α)(CRONBACH 1951) according to the following relationwhere k is the number of test items, ���� 2 is the sum of the item variance; s2 is the variance of the total score:

�� k-1 ×(1-

∑�� ��=1����

2 ��2 ) (1)

The level of Cronbach alpha is 0.75, which means from the point of view of consistency, our questionnaire could be accepted Based on the FINSTAT (2020) data about all enterprises, it was possible to compile a database of WPI enterprises, a total of 5.343 operating in Slovakia. In order to meet the condition for generalization of measured data and obtained results to the whole population of enterprises, it is necessary to meet the minimum sample size (n), which was determined by the following relationship to the calculation under conditions of the permissible error of 5% (e=0.05) and confidence level of 95% (z=1.96) at known base set size (N=5343) and p-level (p=0.5) (KOZEL 2006, FAERON 2017):

��∗(1−��)

��2 ��2+��∗(1−��) �� (2)

Using the formula above, it was possible to calculate the minimum sample size for the survey. The sample must consist of at least 359 respondents. The questionnaire was sent to 2,500 enterprises via their e-mail contacts in the period from September 2020 to the end of that year. As 404 respondents took part in the survey, the results of the survey can be generalized to the whole basic set – the survey meets the condition of a minimum sample size. The results of the questionnaire survey were processed in the Statistica program and the following statistical methods were used to verify the assumed hypotheses: Hypothesis test of relative abundance and Interval estimation of relative frequency. The Hypothesis test of relative abundance is used to test a statistical hypothesis that the proportion of a certain value of a variable in the base set is equal to a given constant according to the following relation:

��−��0

√��(1−��) �� (3)

The Interval estimation of a parameter of the basic set determines the numerical interval in which the estimated parameter occurs with a certain probability (i.e.,the estimated parameter is in the interval (q1,q2) with the probability of 1-α). The interval (q1,q2) is called the confidence interval and depends on α (RIMARČÍK 2007). Hypotheses were verified through Interval estimation according to the relation for the calculation of the 95% Confidence interval for relative frequency:

(4)

The results of the survey were also processed through descriptive statistics and clear graphs were used for their simplicity, comprehensibility, and graphical clarity to display the measured data.

The following hypotheses have been formulated: 125

H1: It is assumed that at least half of the Slovak woodworking and furniture enterprises are family businesses.

Family businesses, as evidenced by their shares of employment (50% to 80%) and GDP (70%) worldwide are irreplaceable and together with their specific advantages and overall share of almost two-thirds of all companies, they represent a pillar of the market economy (BEKERIS 2012, PARADA and GIMENO 2016, WANG et al. 2017). For the conditions of the Slovak business environment, the SBA (2020), based on its own research of the family business issue, presented an estimate which states that approximately 60 to 80% of Slovak enterprises in all sectors are family-type. A specific study in the woodworking and furniture industry connected to the family business has not been published so far. This is one of the reasons why verification and assessment of the conditions are among the first priorities. H2: It is assumed that the main external problem of the development of woodworking and furniture enterprises is the intensity of competition growth from the perspective of other enterprises in the sector. H3: It is assumed that the main internal problem of Slovak woodworking and furniture family businesses is the lack of a qualified workforce and disharmony of the interaction between the elements of family and business.

Every company operating in a market economy is to some extent influenced by both external and internal determinants of the business environment (MISZTAL and KOWALSKA 2020, HERNÁNDEZ et al. 2020). Most factors of the external environment are invariable (tax and regulatory burden, legal and legislative environment, business conditions, etc.). These conditions have a major impact on business, and the European Union, together with the individual Member States, has long been working to eliminate them by seeking to remove obstacles to the development of European businesses by simplifying legislation and improving business conditions (BRAGARU 2015, IONESCU et al. 2011). However, it is also possible to speak of a group of partially variable factors (own activity or attitude towards them) such as competition, while the problem of the SME sector is the relatively low competitiveness, both against large enterprises as well as other enterprises (MALEGA 2017, ABDULAALI et al. 2019). The group of factors that an enterprise creates, modifies and influences is referred to as internal problems. A specific internal problem in the woodprocessing industry is the provision of a qualified workforce, and therefore it can be assumed that the problem also affects family businesses (KOVALČÍK 2018). A particularly specific problem of family businesses is to achieve harmony between family and work life of the family members (ZHOU 2014).

RESULTS AND DISCUSSION

A total of 282 FBs participated in the survey about the current position of family woodworking and furniture enterprises in Slovakia (Table 1). The results show that their share is more than half. To verify the validity of hypothesis H1 that at least half of the enterprises in the woodworking and furniture industry are family-owned, statistical verification was performed by the Hypothesis test of relative abundance with the result of p = 0.000 (Table 2). At the same time, for more detailed data, the results obtained using the Interval estimate of the relative frequency with 95% confidence (Table 3) showed that the share of FBs in the woodworking and furniture sectors is between 65% and 74%. From several available sources, whether domestic or foreign (BEKERIS 2012, PARADA andGIMENO 2016, WANG et al. 2017, SBA 2020), it is possible to deduce the majority shares of family businesses within all currently operating companies worldwide for all sectors. Based on the above results of the statistical verification of the validity of hypothesis H1, it can be

126

concluded that even in Slovakia in the woodworking and furniture sectors, the share of family businesses is major.

Tab. 1 Frequency table of the research sample.

Type of enterprise in the sample Absolute frequency

Cumulative absolute frequency

Family business 282 282 Non-family business 122 404 Total 404 Relative frequency (%)

Cumulative relative frequency (%) 69.80 69.80 30.20 100.00 100

Tab. 2 Hypothesis H1 test on relative abundance.

Hypothesis Research area Alternative hypothesis p n u p-level

H1 Share of FBs in woodworking and furniture industry π > 50% 69.80% 404 8.67 0.000

Tab. 3 Hypothesis H1 Confidence interval for relative frequency. Hypothesis Research area p n

95% Confidence interval Lower limit Upper limit H1 Share of FBs in woodworking and furniture industry 69.80% 404 66% 74%

The questioned family businesses in the woodworking and furniture industry in Slovakia identified the problem factors from the internal and external environment they face and consider them a priority in terms of further development. Figure 1 shows a total of 12 evaluated determinants and their percentage representation. The key internal determinant in the development of woodworking and furniture FBs is the lack of a qualified workforce, which was expressed by almost 77% of respondents. From the results of the Interval estimate for relative frequency with 95% confidence, it can be determined that the given determinant occurs in 72% to 82% of enterprises (Table 5). First part of hypothesis H3 was confirmed by a Hypothesis test of relative abundance result with a p-level value (p = 0.000) (Table 4). Among the verified external determinants, attention was paid to the intensity of competition growth in the woodworking and furniture sectors. The available data for this determinant showed a rate of only 14.89% compared to the determinant of the rising cost, which also belongs to the external group. An estimate from 11% to 19% of FBs (Table 5) in the woodworking and furniture sectors worry about the growing competition, moreover, only 2% to 7% feel the inability to cope with it. In the case of the determinant of rising costs, 60.99% of the respondents expressed a degree of importance, i.e., significantly higher. Hypothesis H2 was not confirmed based on the Hypothesis test of relative abundance (p = 1.000) (Table 4).

127

suppliers problems

inability to face competition

management shorcomings

rising costs

lack of funds

qualified workforce 10.28%

3.90% 4.25% 14.89% 7.09% 10.28%

8.51%

7.44% 33.33%

36.87%

60.99%

76.95%

0% 20% 40% 60% 80% 100%

Fig. 1 Internal and external determinants of family woodworking and furniture enterprises (Source: authors)

Tab. 4 Hypothesis H2 and H3 test on relative abundance.

Hypothesis Research area Alternative hypothesis p n u p-level

H2 Determinant of increasing competition in the industry π > 50% 14.98% 282 -16.56 1.000

H3 Determinant of lack of qualified workforce π > 50% 76.95% 282 10.74 0.000

H3 Disharmony of family and business π > 50% 57.44% 282 2.53 0.006

Tab. 5 Hypothesis H2 and H3 Confidence interval for relative frequency.

Hypothesis Research area p n

95% Confidence interval Lower limit Upper limit H2 Determinant of increasing competition in the industry 14.98% 282 11% 19% H3 Determinant of lack of qualified workforce 76.95% 282 72% 82% H3 Disharmony of family and business 57.44% 282 52% 63%

The question arises, what are the specific problems of family businesses in the woodworking and furniture sectors, assuming that there is a disharmony between the elements of family and business. The values for the selected 7 specifics are shown in Figure 2, where the most significant was the neglect of the family at the expense of the business, which significantly exceeds the others. Second part of hypothesis H3 was confirmed by a Hypothesis test of relative abundance result with a p-level value (p = 0.006) (Table 4). As results of Confidence interval for relative frequency (Table 5) show this problem affects from 52% to 63% of woodworking and furniture family enterprises.

128

redistribution of competences issue of succession insufficient qualification of other employees insufficient qualification of family employees unclear competencies of family employees family neglect conflicts of family employees 9.90%

6.02%

23.04% 17.37%

6.38%

8.15% 57.44%

0% 10% 20% 30% 40% 50% 60% 70%

Fig. 2 Specific problems of family woodworking and furniture enterprises (Source: authors).

The second most common specific of FBs is the solution of the issue of succession. Specifically, 23.04% of respondents consider the subsequent transfer of the company to the next generation to be problematic. The smooth succession of the company to the next generation can be ensured by the succession strategy. Most enterprises in Slovakia (which started their activities after 1993) have not experienced the succession process yet, and according to the results of the Interval Estimate, 76% to 85% do not even have a specific succession strategy. The succession itself occurs approximately every 20 to 25 years and is managed well by only a third (PERÁČEK et al. 2020, VILČEKOVÁ et al. 2018).

In addition to the above facts as the share of FBs in the woodworking and furniture industries in Slovakia, the determinants hindering their development and specifics, the paper seeks to address the areas of future development of the enterprises (Figure 3). The results of the survey clearly showed that the dominant areas in which targeted attention is needed are technical equipment and production technology. The high percentage (68.79%) of respondents, who anticipate future development in any area of the company, is positive.

reaching new customers

quality of production HR offered assortment production technology technical equipment

36.08% 37.11% 34.02% 30.92%

67.52% 70.61%

0% 10% 20% 30% 40% 50% 60% 70% 80%

Fig. 3 Assumed areas of future development of woodworking and furniture FBs (Source: authors).

The above results clearly showed that there is a majority share of family businesses in Slovakia in the woodworkingand furniture industries, as stated by the SBA (2020)forfamily businesses in general, BEKERIS (2012), PARADA and GIMENO (2016), WANG et al. (2017) and PERÁČEK et al. (2020). On the sample of 404 woodworking and furniture enterprises, 282 FBs (69.8%) were identified. The result of the Hypothesis test of relative abundance (p = 0.000) and the Interval estimate for relative frequency with 95% confidence, which determined the range of FB share in the given sectors to 65% to 74%, can validate the hypothesis H1 about the majority share of family businesses in woodworking and furniture sector in Slovakia. The authors MACHEK et al. (2013), SZABÓ (2012), MARTIN (2008) and WRÓBLEWSKA-KAZAKIN (2012) agree that without the adoption of a specific definition in the legislative conditions, the collection and evaluation of data on family business are not possible and at the same time there is no room for its further development. From the point

129

of view of the examined determinants of the internal and external environment, which affect each enterprise in the market economy (MALÁ et al. 2017) for the group of internal determinants, the most important was the provision of a qualified workforce confirmed by a Hypothesis test of relative abundance result with a p-level value (p=0.000), as also expressed by AYMEN et al. (2019), SEDLIAČIKOVÁ et al. (2021), KOVALČÍK (2018). It is estimated by the Interval estimate for relative frequency that 72% to 82% of family woodworking and furniture businesses face this problem. Along with the above, family businesses are also characterized by their specific areas (ZHOU 2014, RAMADANI and HOY 2015). In the research area, the central problem of woodworking and furniture family businesses can be considered the disharmony of family and business elements, i.e., family neglect at the expense of the business, as expressed by up to 57.44% of respondents and confirmed by the a Hypothesis test of relative abundance result with a p-level value (p = 0.006). According to the results of the Interval estimate for relative frequency, 52% to 63% solve the given problem. The second most serious problem is the issue and the process of succession, which in practice is not given sufficient attention, respectively it is neglected (SZABÓ 2012, PETRÁČEK et al. 2020). As many as 80% of the addressed enterprises have not solved the issue of succession yet, while only 16% of enterprises worldwide have a formally established succession strategy and a selected successor (ZAJKOVSKI and DOMANSKA 2019, HAVIERNIKOVA et al. 2019). Succession is considered to be the most critical period of a family business, and founders usually try to delay this moment as much as possible (MURA and KAJSAR 2019). From the findings of the internal determinants, and thus on the problem of securing a qualified workforce and, in the case of the specifics of family businesses, the problem of maintaining harmony between family and business, the H3 hypothesis can be confirmed. Even though the authors MISZTAL and KOWALSKA (2020), HERNÁNDEZ et al. (2020), BRAGARU (2015), IONESCU et al. (2011), MALEGA (2017) and ABDULAALI et al. (2019) agree that competition and ability to face it are considered to be strong external determinants and the hypothesis H2 assumed it, it was not possible to confirm it (according to the result of a Hypothesis test of relative equal p-level value (p = 1.000)). Out of the studied external factors, the determinant of rising costs was identified to be the most significant (60.99 %). The indicator of growing competition (14.98%) that worries family woodworking and furniture business is estimated by the Interval estimate for relative frequency at 11% to 19%, while the indicator of inability to face it (only 4.25%) is estimated at only 2% to 7%.

CONCLUSION

A total of 404 woodworking and furniture enterprises took part in the survey of the current position of family woodworking and furniture businesses in Slovakia. According to the definition of FB, it was possible to clearly identify up to 282 family businesses. The results of the survey show that most enterprises in the woodworking and furniture sector are family-run, with an estimated share ranging from 65%to 74%.These enterprises are affected by many factors that hinder their development. The most important internal determinant is considered to be the provision of a qualified workforce, however, there is a shortage of it. The specifics of family businesses indicate that the neglect of the family at the expense of the business has emerged as a serious problem. It follows from the above that two of the three assumed hypotheses were confirmed. The hypothesis of external determinants was not confirmed, where the assumption was formulated that the most serious is growing competition. In this case, the determined key factor was the rising costs for maintaining the business. The aim of the paper has been met and thus provides an insight into the situation

130

of woodworking and furniture family businesses, their determinants and the results finally also provide an overview of areas for future development. These include investments in technical equipment and production technology.

REFERENCES

ABDULAALI, A., ALNOOR, A., ENAIZAN, B. 2019. A multi-level study of influence financial knowledge management small and medium enterprises. In Polish Journal of Management Studies 190, 21-31. ANTOV, P., SAVOV, V., NEYKOV, N. 2020. Sustainable bio-based adhesives for eco-friendly wood composites a review. In Wood Research 65, 51-62. DOI: 10.37763/wr.1336-4561/65.1.051062 ARREGLE,J.L.,DURAN,P.,HITT,M.,VAN ESSEN,P.2016. Why is family firms’ internationalization unique? A meta-analysis. Entrepreneurship Theory and Practice. DOI: 10.1111/etap.12246 AYMEN, A., ALHAMZAH, A., BILAL, E. 2019. A multi-level study of influence financial knowledge management small and medium enterprises. In Polish Journal of Management Studies 19, 21-31. BEKERIS,R. 2012. The impact of macroeconomic indicators upon SME´s profitability. In Ekonomika 91, 117-128. BRAGARU,C. 2015. Small and medium enterprises (SMEs) – determinants for intelligent growth and sustainable development of local communities. In Challenges of the Knowledge Society 1, 641-651. CRONBACH,L.J.1951.Coefficient alpha and the internal structure of tests. In Psychometrika 16, 297334. DOI:10.1007/BF02310555 DE MASSIS,A.,KOTLAR,J.,MAZZOLA,P.,MINOLA,T.,SCIASCA,S.2016. Conflicting selves: Family owners’ multiple goals and self-control agency problems in private firms. In Entrepreneurship Theory and Practice. DOI:10.1111/etap.1225.7 FAERON, E. 2017. Sample Size Calculations for Population Size Estimation Studies Using Multiplier Methods with Respondent-Driven Sampling Surveys. JMIR Public Health and Surveillance 3, 59. FINSTAT 2020. [WWW Document]. URL https://finstat.sk/databaza-firiemorganizacii?Activity=drevo+a+papier&Region=&SalesFrom=&PerPage=&Sort=creation-datedesc&Tab=. HAJDÚCHOVÁ, I., HLAVÁČKOVÁ, P. 2019. Vplyv globálnej ekonomiky na lesnícko-drevársky sector v Českej a Slovenskej republike. In Acta Facultatis Xylologiae Zvolen 560, 135-146. HALAJ, D., SEDLIAČIKOVÁ, M., MALÁ, D. 2018. Customer Behavior on the Slovakian Roundwood Market: A Case Study. In BioResources 13, 6003-6020. HAVIERNIKOVA, K., SNIESKA, V., NAVICKAS, V., BURKSAITIENE, D. 2019. The attitudes of small and medium entrepreneurs toward cluster cooperation: the expectations and reality. In Transformations in Business & Economics 18, 191-205. HENNART, J. F., MAJOCCHI, A., FORLANI, E. 2019. The myth of the stay-at-home family firm: How family managed SMEs can overcome their internationalization limitations. In Journal of International Business Studies 50, 758-782. DOI:10.1057/s41267-017-0091-y HERNÁNDEZ, Y. G., GALVIS, J. F. R., CAPACHO, J. W. V. 2020. Business survival, a literary review of internal and external determinants. In Criterio Libre 18, 261-279. IONESCU, V., CORNESCU, V., DRUICA, E. 2011. Small and medium enterprises in European Union. In Journal of Business and Leadership 1, 55-67. KOVALČÍK, M. 2018. Efficiency of the Slovak forestry in comparison to other European countries: An application of Data Envelopment Analysis. In Central European Forestry Journal 64, 46-54. KRIESTIE, L. 2002. The World´s Oldest Family Companies. In Family Business Magazine, 61-67. KRIŠŤÁKOVÁ, S., NEYKOV, N., ANTOV, P., SEDLIAČIKOVÁ, M., REH, R., HALALISAN, A., HAJDÚCHOVÁ, I. 2021. Efficiency of Wood-Processing Enterprises-Evaluation Based on DEA and MPI: A Comparison Between Slovakia and Bulgaria for the Period 2014-2018. In Forests 12. DOI:10.3390/f12081026 KOZEL, R. 2006. Moderní marketingový výzkum: nové trendy, kvantitativní a kvalitativní metody a techniky, průběh a organizace, aplikace v praxi přínosy a možnosti, 1 stedn. ed. Grada, Praha.

131

MACHEK, O., HNILICA, J. 2013. On the Performance Gaps between Family and Non-Family Firms in the Czech Republic. In Central European Business Review 2, 54-55.

MALÁ, D., SEDLIAČIKOVÁ, M., KAŠČÁKOVÁ, A., BENČÍKOVÁ, D., VAVROVÁ, K., BIKÁR, M. 2017. Green Logistics in Slovak Small and Medium Wood-Processing Enterprises. In Bioresources 12, 5155-5173. MALEGA, P. 2017. Small and Medium Enterprises in the Slovak Republic: Status and Competitiveness of SMEs in the Global Markets and Possibilities of Optimization. In Optimal Management Strategies in Small and Medium Enterprises, 102-124. MARINOVA,S.,MARINOV,M.2017. Inducing the internationalization of family manufacturing firms in a transition context. In European Business Review 29, 181–204. MARTIN,R. 2008. Post-Socialist Segmented Capitalism: The Case of Hungary Developing Business Systems Theory. In Human Relations 61, 131-159. MISZTAL, A., KOWALSKA, M. 2020. Determinants of sustainable development of industrial enterprises in Poland in the period from 2010 to 2019 – a statistical evaluation. In Prace Naukowe Uniwersytetu Ekonomicznego we Wroclawiu 64, 160-173. MORESOVÁ, M., SEDLIAČIKOVÁ, M., KAŠČÁKOVÁ, A. 2020. Global determinants of sustaining and developing family enterprises in Slovakia. In Globalization and its Socio-Economic Consequences 2019. 74. DOI:10.1051/shsconf/20207403005 MORESOVÁ, M., SEDLIAČIKOVÁ, M., ŠTEFKO, J., BENČIKOVÁ, D. 2019. Perception of Wooden Houses in the Slovak Republic. In Acta Facultatis Xylologiae Zvolen 61, 121-135. DOI:10.17423/afx.2019.61.2.12 MURA, L. 2019. Entrepreneurship Internationalization –Case of Slovak Family Business. In Ad Alta: Journal of Interdisciplinary Research 9, 222-226. MURA, L., KAJZAR, P. 2019. Small Business in Cultural Tourism in a Central European Country. In Journal of Tourism and Services 10. DOI:10.29036/jots.v10i19.110 MUSOVA, Z., MUSA, H., MATIOVA, V. 2021. Environmentally responsible behaviour of consumers: Evidence from Slovakia. In Economics and Sociology 14, 178-198. DOI: 10.14254/2071789X.2021/14-1/12 PARADA, M. J., GIMENO, A. 2016. Understanding Characteristics of Long-Term Value Creating Family Business. In Effective Executive 14, 43-54. PERÁČEK,T.,VILČEKOVÁ,L.,STRÁŽOVSKÁ,Ľ.2020. Selected Problems of Family Business: A Case Study from Slovakia. In Acta Polytechnica Hungarica 1, 145-162. RAMADANI, V., HOY, F. 2015. Context and Uniqueness of Family Businesses. In Springer, Cham. DOI:10.1007/978-3-319-14209-8_2 SBA 2020. Bariéry rodinného podnikania. [WWW Document]. URL http://monitoringmsp.sk/wpcontent/uploads/2020/04/Bari%C3%A9ry-rodinn%C3%A9ho-podnikania-na-Slovensku-3.pdf SCHEER, Ľ. 2007. Biometria. Zvolen: Technická unvierzita vo Zvolene, Lesnícka fakulta, 2007. 333 p. ISBN 978-80-228-1723-3. SCHOLES,L.,MUSTAFA,S.,CHEN,S.2015. Internationalization of small family firms: The influence of family from a socioemotional wealth perspective. In Thunderbird International Business Review 58, 131–146. SEDLIAČIKOVÁ, M., MORESOVÁ, M., ALÁČ, P., DRÁBEK, J. 2021. How Do Behavioral Aspects Affect the Financial Decisions of Managers and the Competitiveness of Enterprises? In Journal of Competitiveness 13, 99-116. SZABÓ, A. 2012. Family business in Hungary. 10th International Conference on Management, Enterprise and Benchmarking, June 1-2, 2012, Budapest, Hungary. ŠEBEŇ, V., KUČERA, M., MERGANICOVA, K., KONÔPKA, B. 2018. The current state of non-forest land in the Czech Republic and Slovakia - forest cover estimates based on the national inventory data. In Forestry Journal 64, 207-222. DOI:10.1515/forj-2017-0043 SHARMA, P. 2004. An Overview of the Field of Family Business Studies: Current Status and Directions for the Future. In Family Business Review 17, 1-36. DOI:10.1111/j.17416248.2004.00001.x VERBEKE, A., FOROOTAN, M. 2012. How good are multinationality–performance (M–P) empirical studies? In Global Strategy Journal 2, 332–344.

132

VILČEKOVÁ,L.,MUCHA,B.,PERÁČEK,T.,STRÁŽOVSKÁ,Ľ.2018. Selected issues of family business in selected countries with emphasis on the Slovak Republic. In Proceedings of the 31st International Business Information Management Association Conference, Milan, Italy, April 26-28, 129-138. WANG, Y., DELGADO, M. S., MARSHALL, M. I., DOBBINS, C. L. 2016. Generational Shadow in Farming Business: How Does It Affect the Succession Process? Annual Meeting, February 4-7, 2017, Mobile, Alabama, Southern Agricultural Economics Association. WRÓBLEWSKA-KAZAKIN, A. 2012. Problemy decyzyjne w firmach rodzinnych naprzykladzie średniego przedsiebirostva. In Przedsiebiorczośći Zarzadzanie 13, 129-142. ZAJKOWSKI, R., DOMAŃSKA, A. 2019. Differences in perception of regional pro-entrepreneurial policy: does obtaining support change a prospect? In Oeconomia Copernicana 10, 359-384. ZHOU, L. 2014. Social responsibility and employees´organizational identification in Chinese family firms. In Chinese Management Studies 8, 683-703.

ACKNOWLEDGEMENT

The authors are grateful for the support of the Slovak Research and Development Agency, grants number APVV-18-0520, APVV-18-0378, APVV-17-0456, APVV-17-0583, APVV-20-0004 and project KEGA 005TUZ-4/2020 also project IPA 09/2021. This publication is also the result of the project implementation: Progressive research of performance properties of wood-based materials and products (LignoPro), ITMS: 313011T720 (10%) supported by the Operational Programme Integrated Infrastructure (OPII) funded by the ERDF.

AUTHORS’ ADDRESSES

Ing. Anna Kocianová (ORCID: 0000-0001-6169-8578) prof. Ing. Mariana Sedliačiková, PhD. (ORCID: 0000-0002-4460-2818) Ing. Mária Moresová, PhD. et PhD. (ORCID: 0000-0001-6815-0724) Technical University in Zvolen Faculty of Wood Sciences and Technology, Department of Business Economics T. G. Masaryka 24, 960 01 Zvolen, Slovakia xkocianovaa@is.tuzvo.sk sedliacikova@tuzvo.sk maria.moresova@tuzvo.sk

Mgr. Jarmila Schmidtová, PhD. (ORCID: 0000-0003-3985-9616) Technical University in Zvolen Faculty of Wood Science and Technology, Department of Matematics and Descriptive Geometry T.G. Masaryka 24, 960 01 Zvolen, Slovakia jarmila.schmidtova@tuzvo.sk

133

134

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 64(1): 135−146, 2022 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2022.64.1.12

CAPITAL STRUCTURE DETERMINANTS OF WOOD-PROCESSING ENTERPRISES IN SLOVAKIA

Peter Krištofík – Juraj Medzihorský

ABSTRACT

Capital structure has been widely discussed, but there is still a lack of industry-oriented empirical studies of Slovak enterprises focusing on this topic. The aim of the paper is to identify significant capital structure determinants of wood-processing enterprises in Slovakia, and their comparison with the determinants in other industries worldwide. Applying panel regression, in the years 2016-2019, we found evidence for the negative relation between leverage and profitability, growth opportunities, cash, respectively. But most of these relations were disturbed during the crisis in 2020. Some evidence of a negative relation between leverage and size occurred only in the crisis period. There is only a partial confirmation of several capital structure theories. Pecking-order theory corresponds to the debt-equity choice of Slovak wood-processing enterprises best. When comparing empirical capital structure determinants with other industries and countries, the most similar to our sample seems to be the food and beverages industry in Indonesia. Our paper is the first one, which reveals relations between leverage and its determinants of Slovak wood-processing enterprises that support a need for next studies focusing on similar topics. Key words: capital structure, leverage, wood-processing enterprises, Slovakia, panel regression

INTRODUCTION

Capital structure theories and their empirical verification on different datasets have belonged to widely discussed topics among economists since MODIGLIANI and MILLER (1958) presented their irrelevance theory of capital structure (original MM model). An inexhaustibility of the topic lies in various theories and in the fact that studies of different countries and industries can have both similar and different results.

Adding only one variable – corporate taxes – to the original MM model, capital structure is relevant thanks to interest tax shield (MODIGLIANI and MILLER 1963). Optimal capital structure would be represented by zero equity. On the other hand, the use of other techniques for tax optimization, like non-interest tax shield (depreciation & amortization), can lower a motivation for higher leverage, as non-interest and interest tax shields are substitutes (TREZEVANT 1992). It is clear that enterprises with minus EBIT are not motivated to use interest tax shield, and enterprises with minus EBITDA do not need any tax shield, while enterprises with relatively high EBITDA do not necessarily consider the shields as substitutes; but as complements.

135

Trade-off theory (KRAUS and LITZENBERGER 1973, MYERS 1984) adds the next variable – financial distress costs. As benefits of interest tax shield must be compared to these expenses, optimal capital structure will not have a corner solution (maximal leverage, zero equity), but the interior one. We can say that business risk is the determinant of possible financial distress that can be amplified by high leverage. Therefore, lower leverage is suitable for enterprises with higher business risk and with higher bankruptcy costs, including the indirect ones such as investment restrictions, a loss of customers, business partners, growth opportunities and key employees (KRIŠTOFÍK 2010), reputational damage of owners and managers that is more severe in specific markets with limited number of potential business partners etc.

Agency costs theory (FAMA and MILLER 1972, JENSEN and MECKLING 1976) focuses on relations between mangers, owners, and creditors. To minimize agency costs of debt, a collateral can be used, owners - especially of micro and small enterprises - can offer personal guarantees, an enterprise as whole or selectively adebt issue should be rated,any information asymmetry between insiders and outsiders should be minimized etc. Minimizing agency costs of equity includes control mechanisms, incentive schemes, and debt issues, especially if cash is so high and investment opportunities low that it motivates managers to an ineffective consumption. Actually, the high value of cash to total assets ratio is the typical feature of many Slovak wood-processing enterprises. Agency costs of both equity and debt can also be minimized thanks to ESG disclosures, ratings, and rankings, with a focus on Ggovernance ones.

Pecking-order theory (MYERS and MAJLUF 1984) supposes that an enterprise follows certain order when financing: internal funds, debt issue, equity issue. Enterprise issues new debt or equity only if internal funds are insufficient for investment opportunities. So, the need for external financing relates on investment opportunities and on internal funds items, flows, respectively: profitability (retained earnings), dividend policy, amount of cash, depreciation and amortization. If we omit the least probable equity issue, the determinants of external financing become the determinants of debt issue and leverage. If we do not omit equity issue, debt-equity choice is not clear, as equity is both the first and the last financing option.

According to life cycle theory (WESTON and BRIGHAM 1981,CHITTENDEN et al. 1996), smaller and younger enterprises can have some constraints with obtaining new external funds. However, those problems can be minimized by a collateral, and other techniques for achieving investors’ trust described in the agency costs theory. The next financing options for start-ups include venture capital, business angels, and internal funds when the enterprises become profitable.

Every rational subject should buy cheaply and sell expensively. Applying this simple idea to capital markets, an enterprise should issue new shares when stock price is relatively higher and make buybacks when it is lower. According to market timing theory (BAKER and WURGLER 2002), this effect can be long-lasting. Therefore, current capital structure is the result of past market timing activities. Applying the theory on debt, there should be a negative relation between interest rates and debt issues.

Empirical studies of capital structure are very often country-oriented, less often industry-oriented. ALMAZAN and MOLINA (2001) studied 61 industries using the Compustat database. They show different leverage variability in different industries. Higher withinindustry variability of leverage is confirmed for industries with a longer history, higher capital expenditures, and greater leas indebtedness. Regarding corporate governance variables, board structure and incentives play a role.

CAPOBIANCO and FERNANDES (2004) studied the enterprises of airline industry over the world applying DEA models. They identified that more effective enterprises have lower

136

values of fixed to total assets ratio, and equity ratio minimally 40%. The enterprises with rising revenues reduce their leverage.

VIVIANI (2008) confirmed that pecking-order theory is more applicable to French wine enterprises than trade-off theory. Applying both single-year linear regression and panel regression on 5-year dataset, he studied profitability, cash, asset turnover, tangibility, noninterest tax shield, age, and growth in sales as leverage determinants. Statistical significance of the selected determinants was confirmed especially when applying single-year approach. Regarding the methodology, several variant measures were defined for both leverage and its determinants. The specific view on the industry was supplemented by the industry subsectors analysis.

ISLAM and KHANDAKER (2014) confirmed profitability and asset tangibility as significant determinants of the Australian enterprises in the mining industry, while the same determinants are insignificant for other (non-mining) enterprises. Such results support a need for industry-oriented research.

That is confirmed by the next study of Australian listed companies, as well. LI and ISLAM (2019) – analysing 20 industries – found that firm-specific determinants of capital structure vary across industries. In addition, they show the significant industry-specific determinants.

ZHANG, CAO, and ZOU (2016) confirmed an overinvestment in the renewable industry in China, especially for the wind and biomass subsectors. The effect of leverage on profitability is confirmed especially for downstream enterprises. That creates a basis for policy makers to minimize the constraints that enterprises have when acquiring new debt.

SALIM and SUSILOWATI (2019) focused on Indonesian listed food and beverages companies. They found that liquidity and enterprise growth are significant capital structure determinants, with the negative impact on leverage, while profitability and enterprise size are insignificant. The effect of leverage on enterprise value is positive but insignificant.

JAWORSKI and CZERWONKA (2021) joined single-industry with multiple countries research, as they studied capital structure of the energy industry in 25 EU countries. While tangibility and enterprise size correlate with leverage positively, profitability and liquidity have a negative impact. Macroeconomic determinants with strong or moderate effect on leverage are GDP growth, protection of stakeholders’ rights, inflation, taxation, degree of capital markets and financial institutions development.

The complex summary of capital structure theories, selected empirical studies, and the methodology of variables – all focusing on leverage determinants – is presented in Tab. 1. That summary will enable us to develop hypotheses and methodology. We would like to stress that all mentioned determinants have their empirical confirmations in plenty of studies – especially industry non-oriented – but not all of them are part of our industry-oriented review.

Wood-processing industry - which consists of wood, furniture, pulp and paper subindustries - has an important potential in the Slovak economy, which is also given by the forest coverage of the country that is 41% (MINISTRY OF A&RD 2021). Focusing more closely on the first subindustry, its total year revenues (in billions of EUR) in 2020 achieved 1.36, expenses 1.32, respectively (STATISTICAL OFFICE 2022). It has the most important status according to the number of employees, over 12 thousand, followed by the furniture subindustry (NATIONAL FORESTRY CENTRE 2022). However, which determinants have a significant impact on capital structure choice of Slovak wood-processing enterprises, has not been studied yet. The variability of capital structure theories and the existence of industryoriented empiric studies in different countries support a need for this study.

The aim of our paper is to identify significant capital structure determinants and to verify the validity of capital structure theories with a focus on wood-processing enterprises

137

in Slovakia, along with a worldwide comparison of the determinants across industries. The object of the study arewood-processing enterprises except the furniture, pulp and paper ones.

Tab. 1 Capital structure determinants – theoretical and empirical views.

Determinant of capital structure Theoretical effect on leverage

Definition

Corporate income taxes (interest tax shield) MM with tax effect, Trade-off Effective tax rate, Nominal tax rate, MILLER’s (1977) tax index

Depreciation and amortization (non-interest tax shield) MM with tax effect, Trade-off, Pecking-order D&A / Total assets, D&A / Sales, Other non-interest tax shields and tax optimization

Profitability MM with tax effect, Trade-off Pecking-order ROA (EBIT/Total assets), ROI, Margin

Risk and bankruptcy costs (especially indirect) Trade-off Risk: Standard deviation of operating CF / Total Assets, Inverse value of rating, Altman’s model; Bankruptcy costs: Uniqueness of products or industry, High specialization of employees

Empirical confirmation (industry, country, relation to leverage)

Wine – France (- or + depending on leverage measure)

Wine – France (-), Mining – Australia (), Energy – EU (-)

Tangibility (collateral), rating, less information asymmetry from creditors’ view Agency-costs theory (focus on agency costs of debt), Life cycle

Cash Agency-costs theory (focus on agency costs of equity) Pecking-order

Growth opportunities Pecking-order Agency-costs, Trade-off

Dividends Pecking-order

Size and age Life cycle

Stock price

Interest rate Market-timing

Market-timing Fixed assets/Total assets, PPE/Total Assets Collateral(any)/Total assets, (Total assets - Intangible assets)/Total Assets, Existence and value of: Overall credit rating, Issue credit rating, ESG score

Cash/Total assets, Liquidity ratios, Free cash flow/Total Assets

Total assets growth, Sales growth, Market-to-book ratio, Capital expenditures/Total assets

Dummy variable (yes/no), Dividend payout ratio

Size: Total assets, Sales, Market capitalization; Age: Number of years Market-to-book ratio, Current price/Average historical prices Effective interest rate, Weighted average costs of debt Wine – France (+), Mining – Australia (+), Energy – EU (+)

Wine – France (-), Food & beverages –Indonesia (-), Energy – EU (-) Airline – World (-), Wine – France (+), Food & beverages –Indonesia (-)

Energy – EU (+)

138

MATERIAL, METHODS, AND HYPOTHESES DEVELOPMENT

Based on theories and empiric studies – summarized in Tab. 1 – we set the hypotheses and methodology of variables, as follows.

H1: There is a positive linear correlation between leverage and tangibility. The hypothesis confirmation is in favour of agency costs and life cycle theory. Opposite correlation than expected does not have direct support in any capital structure theory. However, as equity is long-term capital, and we include both long- and short-term debt to leverage computation, it can be explained as a possible partial application of the golden financing rule - saying that fixed assets should be financed by long-term capital. In that case, future research focusing on long-term leverage is needed to support or reject such interpretation.

Tangibility can be defined as the ratio of fixed assets to total assets, the ratio of property, plant, and equipment (PPE) to total assets, as collateral value, or literally as tangibility i.e., excluding intangible assets from calculation. PPE has naturally the key role in the wood-processing industry, the ratio of PPE to total assets is relatively high in many wood-processing enterprises and PPE, especially real estate, serves as a collateral. In the case of smaller or younger enterprises, real estate can even be the only suitable collateral together with personal guarantees of owners, when acquiring new debt. Therefore, we define tangibility as PPE to total assets ratio.

H2: There is a positive linear correlation between leverage and size. The hypothesis confirmation is in favour of life cycle theory. The opposite correlation than expected would suggest, for example, a lack of internal funds, or a lack of capital deposits by owners in smaller enterprises. But it does not have any support in any theory.

Enterprise size is most often defined as total assets (its book value) or sales. For listed companies, it can be calculated as market capitalization, or as total assets using market value i.e., as the sum of market value of equity and book value of debt (market value of debt is not usually used, or even calculated). As our dataset does not consist of listed companies, this is not our case. We define size as sales.

H3: There is a positive linear correlation between leverage and growth opportunities. The hypothesis confirmation is in favour of pecking-order theory. On the other hand, if the opposite correlation is confirmed, it will be in favour of agency costs and trade-off theory. Growth opportunities are usually defined as sales growth, total assets growth, market-to-book ratio (for listed companies), and capital expenditures scaled by total assets. We define them as total assets growth, as sales are used for size calculation, and possible collinearity of independent variables should be avoided.

H4: There is a negative linear correlation between leverage and non-interest tax shield. The hypothesis confirmation is in favour of MM model with taxes, trade-off, and peckingorder theory. The opposite result would suggest that both interest and non-interest tax shield can be used together. In other words, their positive correlation would also mean that enterprises which use one tax shield, use also the next tax shield, while other companies apply no one. The first group of mentioned enterprises represent ‘maximal tax optimizers’ . Another explanation is that fixed assets are financed with debt, when both depreciation & amortization and leverage rise. Depreciation and amortization scaled by total assets or by sales are often used for the calculation. We apply D&A scaled by sales to minimize collinearity with variable tangibility that would be caused if the D&A to total assets ratio were used.

H5: There is a negative linear correlation between leverage and profitability.

139

The hypothesis confirmation is in favour of pecking-order theory, while the opposite correlation is in favour of MM model with taxes, and trade-off theory. Profitability can be calculated as ROA, ROI, profit margin, respectively. We define it as ROA – EBIT to total assets – which is the most common measure. It is obvious that EBIT application is suitable for capital structure research due to the fact that both interests that relate to leverage directly and taxes – affected by interest tax shield – are not included.

H6: There is a negative linear correlation between leverage and cash. The hypothesis confirmation is in favour of pecking-order theory. The positive correlation would be in favour of agency costs theory. Variable cash can be defined as cash and cash equivalents scaled by total assets, liquidity ratio (current, quick, cash), or as free cash flow scaled, for example, by total assets. We apply the first one, because a part of studied enterprises has surprisingly very high level of cash and equivalents to total assets ratio; so, it is the important item of assets. Moreover, this ratio has clear and simple interpretation. To sum up, the independent variables are tangibility (TAN), size, growth opportunities (GROWTH), non-interest tax shield (DA), profitability (ROA), and cash. The dependent variable is leverage (LEV). Several definitions are used for that, such as total liabilities to total assets, long-term debt to equity, long-term debt to total assets, short-term debt to total assets, and other variations. We define leverage as the total liabilities to total assets ratio that most complexly includes all non-equity items. It represents debt-equity choice from the broadest view. Mathematical notation of the model is, as follows. Standard symbols for regression are used.

LEV = α + β1 TAN + β2 SIZE + β3 GROWTH + β4 DA + β5 ROA + β6 CASH + ε (1)

As we can see, several capital structure determinants from Tab. 1 are disregarded from our further analysis. Corporate income tax is suitable for multi-country studies, dividends and stock price for studies of listed companies. The rest of determinants are selected for the analysis depending on their application in other industry-oriented studies. Summary of methodology for interpretation of the results is showed in Tab. 2.

Dataset consists of wood-processing enterprises in the Slovak Republic except furniture, pulp and paper ones. Industry is represented by NACEcode 16 – Wood-processing and manufacturing of wood products except furniture (NACE 2022). It includes 16.1. Sawmilling and planing of wood, 16.21 Manufacturing (Ma.) of boards and wooden panels, 16.22. Ma. of parquets, 16.23 Carpentry, 16.24 Ma. of wooden containers, 16.29 Ma. of other wooden products. The five-year period (2016-2020) is studied. As pre-crisis and crisis period should be modelled separately, the period is divided into two subperiods: 2016 –2019, and 2020, due to the pandemic situation. Therefore, we have applied both panel regression for the first subperiod and regression of cross-sectional data for the second subperiod. The database is REGISTRY OF ACCOUNT STATEMENTS (2022). According to the database, the total number of wood-processing enterprises with published account statements for the entire selected period and at least 10 employees is 242. Applying random selection, we have selected one third i.e., 81 enterprises. Their sales represent nearly 32 % market share. We consider such a sample as representative. Micro enterprises (with 9 and fewer employees) are not included, as there are several specifics that make them incomparable with bigger enterprises. Micro enterprises often create account statements only formally, as a necessity required by state – tax office; they include one-person companies that are actually self-employed persons, with only a legal form of a limited company; these

140

enterprises can acquire loans with different conditions than others like EaSI (EU Program for Employment and Social Innovation) loans with guarantee of European investment fund; they often do not follow hardly any governance rules, etc. STATISTICAL OFFICE (2022) uses a partially similar methodology when creating DataCube - only enterprises with 20 and more employees are there analysed exhaustively.

Tab. 2 Methodology for results interpretation (dependent variable - leverage).

Independent variable Regression result Interpretation in favour of … Regression result

Tangibility Agency costs, Life cycle

Size

Growth opportunities

Non-interest tax shield

Profitability Cash β > 0 Life cycle

Pecking-order Maximal tax optimization, Nonexistence of tax shields exhaustion, Fixed assets financed by debt MM with tax effect, Trade-off Agency costs β < 0 Interpretation in favour of … Questionable, possibly golden financing rule with a need for future verification Lack of internal funds and owner’s capital deposits in small enterprises together with too many investment opportunities, Debt preference in debtequity choice in smaller enterprises –unsupported by theories Agency costs, Trade-off MM with tax effect (if tax shields are considered as supplements to each other), Trade-off, Pecking-order Pecking-order Pecking-order

RESULTS AND DISCUSSION

Before the presentation of the final results, let us mention some important results of econometric tests and the reasons for the model adjustments. First, we will look at the panel model. Two high leveraged enterprises were excluded as outliers to achieve an asymptotic normality of the dependent variable and residuals (see Tab. A.1 in Appendix). As our data are burdened with multicollinearity (see Tab. A.2 and A.3), we had to exclude the variable TAN which has the highest pair correlations with other independent variables. According to the Durbin-Watson statistics (value 1.2), autocorrelation also occurs. Therefore, we applied the White method for covariance and standard errors matrix in the final model, which should eliminate both autocorrelation and heteroskedasticity, if any occurs. All variables do not have a unit root (see Tab. A.4) means that their stationarity is not rejected. As many authors prefer scaling the size of an enterprise with logarithm, we applied for SIZE both lin-lin and lin-log models, using natural logarithm. However, the results look very similar (see Tab. 3 and Tab. 4). Random effects are used instead of fixed affects, according to the Hausman test (see Tab. A.5).

When looking at cross-sectional data in 2020, heteroskedasticity occurs according to the White test (see Tab. A.6). Therefore, we applied Huber-White-Hinkley standard error and covariance method consistent with that. Normality of residuals cannot be rejected (see 141

Tab. A.1). Three outliers have been excluded. Similarly to the panel model, both lin-lin and lin-log models – using natural logarithm – are used for variable SIZE. Ramsey RESET test shows correct model specification for the lin-lin model only at 1% significance level, while the lin-log model seems to be even better specified (see Tab. A.7 and Tab. A.8). On the other hand, the log-lin model would not be correct for both panel and cross-sectional data, as residuals would not have a normal distribution.

As we can see in Tab. 3, profitability, growth opportunities, and cash seemed to be relevant capital structure determinants in pre-crisis period. A negative influence of both profitability and cash on leverage is in favour of pecking-order theory. But a negative influence of growth on leverage is against that; and is in favour of agency costs theory. Agency costs theory, however, expects the association of cash and leverage to be opposite to that measured. ROA with minus effect also supports life-cycle theory and is against MM model with tax effect and trade-off theory. These theories are also not confirmed due to the insignificance of DA. As the positive influence of size on leverage is insignificant, it does not support the life-cycle theory. As a result, pecking-order theory seems to describe the capital structure and its relations in the first subperiod better than other theories.

Size became statistically significant in 2020, but its negative correlation with leverage is not in accordance with any theory. Such result can be justified with several explanations. We can deduce for example a lack of internal funds and owner’s capital deposits in small and medium sized wood-processing enterprises that was even confirmed in the National program for the utilization of wood potential in the Slovak Republic (MINISTRY OF A&RD 2013). As this result occurs only in pandemic time, it can also indicate that reduction in equity – that can happen in any crisis – can be more severe in smaller enterprises. However, we did not find the evidence for the second interpretation, as there is only a weak correlation between yearly change of leverage in 2020 and size of an enterprise. Profitability and growth opportunities were not significant determinants of leverage during pandemic year. It can be explained by a reduction in profitability and growth opportunities, which is expectable during any crisis, as values of both variables changed on average in 2019 and 2020 comparing to the rest of the period. On the other hand, the role of cash as capital structure determinant was confirmed again and was even more significant in 2020 than before. This partially supports pecking-order theory. Moralization effect of debt on managers i.e., motivation not to spend disposable cash ineffectively, is therefore not confirmed or needed in the analysed enterprises in any subperiod. Non-interest tax shield was insignificant in both subperiods. Moreover, its positive correlation with leverage is not in accordance with any theory. Therefore, the question also is whether it should be considered as the substitute for interest tax shield. If an enterprise does not meet with tax shields exhaustion, the shields do not have to be substitutes. To sum up the second subperiod, the pecking-order theory is confirmed only partially, and previous relations between leverage and ROA, growth opportunities, respectively, cannot be confirmed.

Tab. 3 Capital structure models. Lin-log model applied for variable SIZE.

Dependent variable: Leverage

Period 2016-2019

2020 Independent variables Beta-coefficient Significance Beta-coefficient Significance LOG(SIZE) 0.012940 -0.055265 ** ROA -0.508273 *** -0.095323 GROWTH -0.003715 *** 0.001611 DA 0.016903 0.005858 CASH -0.281071 ** -0.740272 *** Intercept 0.521946 * 1.524875 ***

142

Note: *, **, *** represents statistical significance according to p-value at the 10%, 5%, and 1% level, respectively.

Important limitation of these comparisons is that the first model is based on 4-year panel data, while the second one only on one-year data.

If we do not apply logarithmic transformation for variable SIZE (see Tab.4) the difference is that SIZE is significant only at 10% level in 2020, not 5%. Positive correlation in the first subperiod in the first model changed to the negative one in the second model, but it is still insignificant.

Tab. 4 Alternative capital structure models. Lin-lin model applied for all variables.

Dependent variable: Leverage

Period 2016-2019 2020

Independent variables Beta-coefficient Significance Beta-coefficient Significance

SIZE -1.14 . 10-09 -2.67 . 10-09 *

ROA -0.488647 *** -0.092732

GROWTH -0.003367 *** -0.012408

DA 0.005537 0.235818

CASH -0.292377 ** -0.667465 ***

Intercept 0.713054 *** 0.737256 *** Note: *, **, *** represents statistical significance according to p-value at the 10%, 5%, and 1% level, respectively.

Capital structure determinants of wood-processing enterprises in Slovakia can be compared to the determinants confirmed in other industries and countries, especially when focusing on the 2016-2019 model. Slovak wood-processing enterprises are similar to French wine enterprises (VIVIANI 2008) regarding profitability, and cash, but growth opportunities show opposite results. There is also a similarity with the mining industry in Australia regarding profitability (ISLAM and KHANDAKER 2014). The results for profitability and cash show some similarity to the energy industry in the EU. However, size is an important determinant in that industry with a positive relation to leverage (JAWORSKI and CZERWONKA 2021) that is not confirmed in our sample. Growth opportunities and cash as leverage determinants with minus correlation are typical for the food & beverages industry in Indonesia. Moreover, size is not confirmed as statistically significant in those enterprises (SALIM and SUSILOWATI 2019), so we can find several similarities there with Slovak woodprocessing enterprises. Regarding growth opportunities, we can find some linkage with the world airline industry, as well (CAPOBIANCO and FERNANDES 2004).

CONCLUSION

In the years 2016-2019, we found the negative relation between leverage and profitability, growth opportunities, cash, respectively. Except for the variable cash, these relations were not confirmed in 2020. Changes in profitability and lack of growth opportunities during crisis thus must have had the impact. Some evidence of a negative relation between leverage and size occurred only in the crisis period. The results are more in favour of the pecking-order theory, than in favour of other theories, but no theory is supported unequivocally. Some similarities between the wood-processing industry in Slovakia and other industries in other countries can be found regarding capital structure determinants. The best example seems to be the food & beverages industry in Indonesia. Regarding the hypotheses, only H5 and H6 are confirmed. Other hypotheses are not 143

confirmed because of the opposite than expected relation, or its statistical insignificance. Practical implications of the results lie for example in some recommendations for such wood-processing enterprises which set their target leverage only as the average of the industry. As the study reveals ‘average’ relations of leverage and its determinants, enterprises without complex capital structure targets can consider these relations rather than following only the industry median leverage.

REFERENCES

ALMAZAN, A., MOLINA, C.A. 2001. Intra-Industry Capital Structure Dispersion. In Journal of Economics & Management Strategy, 14 (2): 263–297. DOI: 10.1111/j.1530-9134.2005.00042.x BAKER, M., WURGLER, J. 2002. Market Timing and Capital Structure. In The Journal of Finance, 57(1): 1–32. DOI: 10.1111/1540-6261.00414 CAPOBIANCO, H. M. P., FERNANDES, E. 2004. Capital structure in the world airline industry. In Transportation Research Part A, 38: 421–434. DOI: 10.1016/j.tra.2004.03.002 CHITTENDEN, F., HALL, G., HUTCHINSON, P. 1996. Small Firm Growth, Access to Capital Markets and Financial Structure: Review of Issues and Empirical Investigation. In Small Business Economics, 8: 59-67. FAMA, E. F., MILLER, M. H. 1972. The Theory of Finance. Holt, Rinehart and Winston: New York. NATIONAL FORESTRY CENTRE. 2022. Forestportal - Economic indicators of wood-processing industry. https://gis.nlcsk.org/IBULH/SpracDrev/DSP [1.2.2022]. ISLAM, S.Z., KHANDAKER, S. 2014. Firm leverage decisions: Does industry matter? In North American Journal of Economics and Finance, 31: 94-107. DOI: 10.1016/j.najef.2014.10.005 JAWORSKI, J., CZERWONKA, L. 2021. Determinants of Enterprises’ Capital Structure in Energy Industry: Evidence from European Union. In Energies, 14, No. 1871. DOI: 10.3390/en14071871 JENSEN, M., MECKLING, W. 1976. Theory of the Firm: Managerial Behaviour, Agency Costs and Ownership Structure. In Journal of Financial Economics, 3: 305–360. KRAUS,A.LITZENBERGER, R.H. 1973. A State‐Preference Model of Optimal Financial Leverage. In The Journal of Finance, 28 (4): 911-922. KRIŠTOFÍK, P. 2010. Finančná ekonómia s aplikáciou na rozvíjajúce sa trhy. Iura Edition, Bratislava. LI,L.,ISLAM, S. Z. 2019. Firm and industry specific determinants of capital structure: Evidence from the Australian market. In International Review of Economics and Finance, 59: 425-437. DOI: 10.1016/j.iref.2018.10.007

MINISTRY OF AGRICULTURE AND RURAL DEVELOPMENT OF SLOVAK REPUBLIC. 2013. National program for the utilization of wood potential in the Slovak Republic. https://www.mpsr.sk/narodnyprogram-vyuzitia-potencialu-dreva-slovenskej-republiky/913-37-913-7913/ [2.4.2022].

MINISTRY OF AGRICULTURE AND RURAL DEVELOPMENT OF SLOVAK REPUBLIC. 2021. Forest coverage has been growing in Europe for a long time. https://www.mpsr.sk/aktualne/skvela-spravalesnatost-v-europe-dlhodobo-stupa/16244/ [5.2.2022]. MODIGLIANI, F., MILLER, M. H. 1958. The Cost of Capital, Corporation Finance, and the Theory of Investment. In American Economic Review, 48, 261–297. MODIGLIANI, F., MILLER, M. H. 1963. Corporate Income Taxes and the Cost of Capital: A Correction. In American Economic Review, 56, 333–391. MYERS,S.C.,&MAJLUF,N. S. 1984. Corporate financing and investment decisions when firms have information that investors do not have. In Journal of Financial Economics, 13 (2), 187-221. MYERS, S. C. 1984. The Capital Structure Puzzle. In The Journal of Finance, 39, 574-592. NACE. 2022. NACE code 16. http://www.nace.sk/nace/16-spracovanie-dreva-a-vyroba-vyrobkovz-dreva-a-korku-okrem-nabytku-vyroba-predmetov-zo-slamy-a-pruteneho-materialu/ [8.2.2022]. REGISTRY OF ACCOUNT STATEMENTS. 2022. https://www.registeruz.sk/cruz-public/domain/ accountingentity/search [10.2.2022]. SALIMM M.N.,SUSILOWATI, R. 2019. The effect of internal factors on capital structure and its impact on firm value: empirical evidence from the food and baverages industry listed on Indonesian stock exchange 2013-2017. In International Journal of Engineering Technologies and Management Research, 6 (7), 173-191. DOI: 10.29121/ijetmr.v6.i7.2019.434 144

STATISTICAL OFFICE. 2022. Data Cube. Selected financial indicators of non-financial corporations. http://datacube.statistics.sk/#!/view/sk/VBD_SLOVSTAT/fp2003rs/v_fp2003rs_00_00_00_sk [7.2.2022]. TREZEVANT, R. 1992. Debt Financing and Tax Status: Tests of the Substitution Effect and the Tax Exhaustion Hypothesis Using Firms' Responses to the Economic Recovery Tax Act of 1981. In The Journal of Finance, 47 (4), 1557-1568. VIVIANI. 2008. Capital structure determinants: an empirical study of French companies in the wine industry. In International Journal of Wine Business Research, 20 (2): 171-19. WESTON, J. F., BRIGHAM, E. F. 1981. Managerial Finance, 7th ed. Hinsdale, Dryden Press. ZHANG,D.,CAO,H.,ZOU, P. 2016. Exuberance in China's renewable energy investment: Rationality, capital structure and implications with firm level evidence. In Energy Policy, 95 (C) : 468–478. DOI: 10.1016/j.enpol.2015.12.005

ACKNOWLEDGEMENT

This paper has been supported by the Scientific Grant Agency of Slovak Republic under the project VEGA No. 1/0579/21 Research on Determinants and Paradigms of Financial Management in the Context of the COVID-19 Pandemic.

APPENDIX

Tab. A.1 Normality of residuals.

Model / Indicator Panel lin-lin Panel lin-log for SIZE 2020 lin-lin 2020 lin-log for SIZE

Skewness

-0.1633 -0.1559 0.0352 0.1387 Kurtosis 2.6567 2.6807 3.1799 3.1890 Jarque-Bera 2.9564 2.6232 0.1213 0.3660 p-value 0.2280 0.2694 0.9411 0.8327

Tab. A.2 Multicollinearity test of 2016-2019 data – correlation matrix.

TAN SIZE ROA GROWTH DA CASH TAN 1 0.2750 -0.2515 -0.0275 0.1589 -0.4710 SIZE 0.2750 1 0.0494 -0.0274 -0.0161 -0.1457 ROA -0.2515 0.0494 1 0.0066 -0.0809 0.2597

GROWTH -0.0275 -0.0274 0.0066 1 -0.0123 -0.0231 DA 0.1589 -0.0161 -0.0809 -0.0123 1 -0.0501 CASH -0.4710 -0.1457 0.2597 -0.0231 -0.0501 1

Determinant of the matrix 0.621542 Farrar & Glauber - X 148.2927 p-value <0.0001 Degrees of freedom 21

Tab. A.3 Multicollinearity test of 2020 data – correlation matrix.

TAN SIZE ROA GROWTH DA CASH TAN 1 0.3156 -0.1768 -0.2084 0.5739 -0.4866 SIZE 0.3156 1 0.0873 -0.0024 0.0153 -0.1714 ROA -0.1768 0.0873 1 0.0961 -0.0879 0.3147

GROWTH -0.2084 -0.0024 0.0961 1 -0.2124 -0.0925 DA 0.5739 0.0153 -0.0879 -0.2124 1 -0.2259 CASH -0.4866 -0.1714 0.3147 -0.0925 -0.2259 1

Determinant of the matrix 0.337548 Farrar & Glauber - X 80.1865 p-value <0.0001

145

Degrees of freedom 21

Tab. A.4 Stationarity tests. Null Hypothesis: Unit root. P-values in the table.

Method/Variable LEV SIZE ROA GROWTH DA CASH

PP - Fisher Chi-square <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

PP - Choi Z-stat 0.0034 0.0006 - <0.0001 0.0144 -

Tab. A.5 Hausman test. Cross-section random effects.

Test Summary Chi-Sq. Statistic Chi-Sq. degrees of freedom

Cross-section random 0.000000 5 P-value

1.0000

Tab. A.6 White test for 2020 model. Null hypothesis: Homoskedasticity.

F-statistic 17.04961 Prob. F(5,72)

0.0000 Obs*R-squared 42.28573 Prob. Chi-Square(5) 0.0000 Scaled explained SS 39.27062 Prob. Chi-Square(5) 0.0000

Tab. A.7 Ramsey RESET Test for lin-log 2020 model. Omitted Variables: Squares of fitted values.

Specification: LEV LOG(SIZE) ROA GROWTH DA CASH C Value Degrees of freedom P-value

t-statistic 1.413219 71 0.1620

F-statistic Likelihood ratio 1.997187 2.163800 (1, 71) 1 0.1620 0.1413

Tab. A.8 Ramsey RESET Test for lin-lin 2020 model. Omitted Variables: Squares of fitted values.

Specification: LEV SIZE ROA GROWTH DA CASH C. Value Degrees of freedom P-value

t-statistic 2.024602 71 0.0467

F-statistic Likelihood ratio 4.099011 4.377946 (1, 71) 1 0.0467 0.0364

AUTHORS’ ADDRESSES

prof. Ing. Peter Krištofík, Ph.D. Ing. Juraj Medzihorský Matej Bel University in Banská Bystrica Faculty of Economics Tajovského 10 975 90 Banská Bystrica Slovakia peter.kristofik@umb.sk juraj.medzihorsky@umb.sk

146

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 64(1): 147−158, 2022 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2022.64.1.13

DETERMINATION OF VARIABLES FOR SOFT SAWNWOOD DEMAND MODELS

Marek Hlodák – Hubert Paluš – Alena Rokonalová – Branko Glavonjić –Katarína Slašťanová

ABSTRACT

The issue of input variables for demand modelling in the soft sawnwood market is addressed in the paper. The procedure is based on theoretical assumptions and possible effects of factors and on the analysis of conditions on wood and wood products market, between which there are certain links. An overview of the theoretical aspects of derived demand, the characteristics of demand on the wood and wood product markets is provided in the introductory part. It is also focused on the determination of market factors. The procedure and selection of input variables for classification into models, their quantification and methods of logical and statistical verification are defined. Basic data and variables for econometric models are captured and used to create models of domestic demand for soft sawnwood. The results represent the basic functional relationships between the factors that affect domestic demand. The benefits relate to the development of knowledge in the field of modelling of the market for wood products and represent a concretization of relevant factors of demand for soft sawnwood in the years 1990-2020 in the Slovak Republic. The basic factors that appear to be suitable for explaining the development of domestic demand for this commodity include gross domestic product, population size, number of completed dwellings, the value of construction production and the price of soft sawnwood. Key words: derived demand, soft sawnwood, demand models, correlation coefficient

INTRODUCTION

Modelling the relationships between market elements is a complex process which involves modelling the supply and demand side of the market and at the same time is the basis for further development of strategies at the national and corporate level (HATRÁK 2007). Demand and supply as the basic determinants of the market are interconnected and regulated by the market price and, retrospectively, the mutual relations between them affect this price (GOLDBERGER 1964). The theory of consumer demand assumes that the demand for a certain product comes from the consumer (satisfaction of needs) and is limited by his/her possibilities (disposable income) (HOLMAN 1999). The quantity demanded is a function of product price, consumer income, prices of substitute and complementary products and consumer preferences (SMRTNÍK 1996). Wood represents one of many inputs to the production process, so that together with other production factors they are transformed into a certain number of outputs (SOLBERG and MOISEYEV 1997). The demand for wood and wood products depends on the demand for final products and is proportional to the activity 147

of industries and final consumers or other producers who use wood and wood products as a source of production process to achieve the final production. In other words, the final demand on the woodmarket depends on the resulting demand for final woodproducts, where the final products are realized on the consumer market. (BAUDIN and BROOKS 1995; BUONGIORNO 1977; GOLDSTEIN and KHAN, 1985). In short, demand for wood products correlates with economic growth (BAUDIN and KANGAS 2003, BUONGIORNO 1978), growth of the construction sector (HURMEKOSKI 2015, BORZIKOWKSI 2017), prices of substitute products (ANYIRO 2013, BAUDIN and KANGAS 2003), with preferences in the use of products (BRÄNNLUND 1988, MICHINAKA 2011), demographic development (BAUDIN and BROOKS 1995, BUONGIORNO 1977) and exchange rate developments (HURMEKOSKI 2015, MICHINAKA 2011). In the broadest sense, demand is diverted from the development of the overall national economy (PALUŠ 2002, HURMEKOSKI 2015, BORZIKOWKSI 2017). In addition to traditional socio-economic factors, the formation of forestry and other related policies, which mainly affect trade, the market, sustainable development, the environment, etc., has an increasingly important influence on the development of the market and trade in wood and wood products (IHEKE 2012, ONOJA 2015, HURMEKOSKI 2013). The country's overall economic growth is the most important factor influencing demand at the wood market (BAUDIN and KANGAS 2003, BUONGIORNO 1978). The size of the economy's output, which can be expressed in terms of gross domestic product (GDP), depends on the development of economic growth. In other words, the growth of GDP is as important as its actual level (LISÝ et al. 2011). If the dependences between the development of the wood market and GDP are known, it is possible to determine the possible development of the wood market on the basis of forecasts of the future development of GDP growth (ESALA et al. 2012).

Developments in the construction sector, mainly fixed capital formation, have a direct impact on the market for mechanical wood processing products - sawnwood and wood based panels (O'CONNOR et al. 2004, ESALA et al. 2012). Activity in the construction sector is very sensitive to changes in economic growth of the country. At the same time, it is a sector where these changes occur as one of the first sectors of the national economy (BUONGIORNO 2009, CEI-BOIS 2004). The size of the disposable income of the population and interest rates has an impact on the construction decision of the population and the availability of investment funds (BAUDIN and KANGAS 2003, BUONGIORNO 1978). Rising prices of construction work and building land reduce construction activity (ESALA et al. 2012, IHEKE 2012). The country governments' program statements and housing policy objectives imply support for housing construction by the government by providing construction premiums, long-term loans from housing development funds, or by supporting the availability of mortgage loans (LUNDMARK 2010). In some countries, new construction (residential and non-residential) plays a key role (MAHAPATRA and GUSTAVSSON 2008), in others the repair and reconstruction sector make up the majority of construction output (O´CONNOR et al. 2004). The volume of wood used in each sector depends on the traditions and uses of the wood (ANYIRO 2013, BAUDIN and KANGAS 2003). Some other sectors that use a significant share of wood products include e.g., industrial production, automotive industry, shipbuilding industry, etc. (PALUŠ 2013). Consumption decisions are also influenced by the existence of alternative options - the availability and price of substitute products on the market. The availability of potential substitution products has an impact on demand elasticity, both in the short and long term. At this level, demand is almost perfectly inelastic. Market information in the field of substitute products or design influences the choice of products and services (EASTIN et al. 2001, MUTANEN 2006, SATHRE and O´CONNOR 2010).

All markets are shaped by the general or individual preferences of end customers (BAUDIN 2003). This phenomenon is partly shaped by culture and partly influenced by the

level of information and knowledge about products and services (BUONGIORNO 2009). Communities use wood and wood products differently, depending on their preferences and traditions in the use of wood raw material (ESALA et al. 2012). The impact of demographic change on the wood market is reflected in increased population pressure on the use of natural resources (O'NEILL et al. 2010). The impact of changes in population structure is therefore important (HETEMÄKI 2011). The structure of the population aims to increase the share of the non-productive age group at the expense of productive people. In conditions of stable economic development, the demands of older people for social security, independent living, etc. will grow (BAUDIN 2003, BUONGIORNO 2009).

The main aim of this paper is determining a set of variables for demand modelling in the soft sawnwood market in Slovakia.

MATERIAL AND METHODS

The selection of input data collection and processing was relatively difficult, especially due to the poor availability of data and their high variability when they were obtained from different sources. Where possible, data from official statistical sources (Statistical Office of the Slovak Republic) were preferred to estimate and recalculate. Input data were drawn from FAOSTAT databases (FAOSTAT 2022), Statistical Office of the Slovak Republic (ŠÚ SR 2022), from the data of the Ministry of Agriculture and Rural Development of the Slovak Republic (MPRV SR 2020) – the Report on Forestry in the Slovak Republic. Variables are expressed in absolute (physical and monetary) units. In order to eliminate the effect of inflation on the values of the input variables and the results obtained, all variables expressed in monetary units, such as GDP or the value of production in the sector are given in constant prices in 2015. Input data form two basic groups of variables – explanatory and dependent variables.

Dependent variables

In the demand models for soft sawnwood, the resulting demand is expressed as domestic consumption of the product i in the year t: Sit = Pit + Iit – Eit (1)

where: Sit - consumption of the product i in the year t, Pit - production of the product i in the year t, Iit - import of the product i in the year t, Eit - export of the product i in the year t. Consumption calculated this way is called apparent consumption and does not take into account changes in stocks in a given year, which differ from actual consumption. Such an approach is not flawed unless the changes in stocks are large and are randomly distributed during the period considered. If annual inventory changes are significant, such an approach can cause large errors in the resulting consumption values and, consequently, errors in the estimated model parameters. If the prices of wood products are derived from the prices of finished products, it is likely that the size of stocks will be closely linked to the development of overall economic activity. For instance, if sales decrease during a period of low economic growth and recession, inventory will increase and the resulting consumption value calculated on the basis of (1) will be overestimated. Despite these complications, in the following analyses we considered data on apparent consumption to be the data on actual consumption. The development of consumption in the years 1990-2020 was analysed and the consumption of soft sawnwood was expressed in m3 .

Explanatory variables

Based on the theoretical assumptions about the development of demand for wood products, we gradually analysed the impact of the following explanatory variables on the size of demand: • dwellings completed, • GDP, • GDP per capita, • construction production, • absolute prices of soft sawnwood.

Non-price variables

The number of completed dwellings expresses the total number of dwellings completed in a given year. The number is given in physical units. The development of the number of completed dwellings in the Slovak Republic during the years 1990-2020 is shown graphically in the results of this paper. GDP is an aggregate indicator that expresses the value of total output and services produced in a given country per year. For the purposes of quantification of models, we used the values of real gross domestic product, which is expressed in billion EUR at constant prices in 2015. GDP per capita is obtained as a share of real GDP and population size in a given year. The value of GDP per capita is expressed in EUR at constant prices in 2015. The absolute values of real GDP and GDP per capita are shown graphically in the results. Construction output includes construction, rebuilding, extension, renewal, repair and maintenance of buildings, including building assembly work and the value of built-in material, carried out by the contractor, own capacities or by subcontracting construction products from other building or non-building organizations for a given year. The value of construction output is expressed in billion EUR at constant prices of 2015. The absolute values of the construction output are shown graphically in the results.

Price variables

The process of obtaining and adjusting price data is limited by their unavailability and non-existent statistical sources on the price development of the main wood products in the Slovak Republic during the period under review. The analysis of the impact of prices on the size of demand requires data on the development of own prices of wood products, the prices of their main competing materials and the prices of final products in the end-use sectors of wood.

The issue of lack of information on price development on the domestic market was solved by approximating domestic market prices of the main wood products using average unit prices of foreign trade, which are calculated from the value and volume of exports or imports of soft sawnwood. The average export and import prices do not represent the actual domestic price. The theoretical assumption about the export and import price is that the export price should be higher and the import price lower than the domestic market price. In terms of free international trade, it can also be assumed that the export price reflects the existing production conditions of the exporting country (input prices, wages, capital price, etc.) and the import price of the production conditions of the country of origin (SMRTNÍK 1992). International trade in wood products is in many cases limited by the passive and active autonomous measures of the state's foreign trade policy, which in turn has an impact on the level of the price. In addition to these barriers, the price in foreign trade is affected by the amount of transport costs, costs of handling goods, insurance of goods, etc. In case data on the price development of soft sawnwood on the domestic market is not available, export or import prices are often used as a substitute for domestic prices (BAUDIN and LUNDBERG 1984, BUONGIORNO 1977, 1978, SCHWARZBAUER 1990). If the volume of exports prevails over imports in a given country, the export price is preferred to the import price and vice

versa. The volume of exports significantly exceeded imports during the examined period. The unit export price therefore relates to a larger volume of soft sawnwood and was used as a substitute for the domestic price for demand analysis. On the other hand, we can assume that the export prices may not be the most appropriate variable to express price conditions on the domestic softwood market. The increase in soft sawnwood production during the period observed may be more strongly motivated by the existing price differences between the domestic and foreign price levels than by the growth in demand and impulses on the domestic market. The domestic price starts to adjust to the export price only with a certain time delay. Given these assumptions, it is possible that at a given point in time, the difference between the export and domestic price of soft sawnwood is higher than the difference between the import and domestic price, thus the import price can be considered a better approximation of domestic prices. Nevertheless, due to the fact that the import of soft sawnwood is approximately 3 to 4 times lower in terms of trade balance, we used the export prices of soft sawnwood.

Correlation analysis methods, graphical methods and other methods of statistical analysis were used to analyse the interdependence of the development of indicators. An initial examination of the relationships between the variables was performed using a scatter plot and a description of their relationship resulting from the graph. Extreme or typical values, possible form of dependencies were determined and the results of the analysis were compared and presented. After an initial graphical review, the phase of searching for exact statistics that confirms the estimates from the graphs has begun. Statistical correlation analysis tools were used for this purpose. It was determined whether there is a relationship between the variables and, if so, what its strength is. The evaluation of the dependence of two random variables is dealt by a simple correlation analysis, which emphasizes more on the intensity of the relationship than on the study of variables in the cause-effect direction (regression). The dependencies we examined were mainly linear, where correlation is a measure of a linear relationship. The important fact is that correlation is not causality. The task of correlation analysis is to identify, quantify and statistically test correlation.

A necessary part is a logical analysis of the problem, in terms of the significance of the correlation itself, which may be distorted or may not exist at all (HENDL 2004). Based on a theoretical review of the functioning of the wood products market, it is possible to define certain assumptions about the relationships between variables. In direct relation to the growing values of one variable, there is an increase in the values of other variable (e.g., the growth of demand has a positive effect on GDP growth). In an indirect relationship with the rising values of one variable, the values of the other variable decrease (e.g., the decline in demand is caused by rising prices). The relationship is uncorrelated if there is no direct or indirect linear relationship between the values of the two variables. In the case of non-price variables, the dependence is assumed to be positive, thus the correlation coefficient will acquire positive values (BAUDIN and BROOKS 1995, BUONGIORNO 1977, GOLDSTEIN and KHAN 1985, HURMEKOSKI 2015, BORZIKOWKSI 2017). As for the price of soft sawnwood, we assume that the dependence will be negative, thus the correlation coefficient will acquire negative values (BAUDIN and LUNDBERG 1984, BUONGIORNO 1977, 1978, SCHWARZBAUER 1990).

The correlation coefficient, like covariance, is a measure of the "mutual difference" of two measured quantities. Unlike covariance, the correlation coefficient is scaled, which means that its value does not depend on the units in which the two measured quantities are given. The value of each correlation coefficient must be from the interval (-1,+1). The analytical correlation tool was used to analyse each mutual combination of measured quantities, which is used to determine the dependence of two measured quantities, i.e., whether higher values of one quantity are related to higher values of the other quantity

(positive correlation), or whether lower values of one quantity are related to higher values of the other quantity (negative correlation), or whether the values of both quantities are independent (correlation close to zero).

The output of the analysis in the form of a table is a correlation matrix, in which the values of the correlation coefficient calculated using Excel 2019 weredisplayed. A graphical representation of the relationship between the explanatory and dependent variables is presented in Fig. 1-6.

RESULTS AND DISCUSSION

Tab. 1 shows the development of selected production and trade indicators of soft sawnwood in the Slovak Republic in the period 1990-2020. Consumption of soft sawnwood is calculated as the production + import - export. Absolute prices are calculated as export prices in EUR, export quantities in m3 . GDP is calculated using the expenditure method at the reference year 2015. Production of sawnwood was the largest before the global crisis in 2008, namely 2,062,861 m3. Consumption has increased by 301,245 m3 and doubled since 1993, the year of the establishment of the Slovak Republic.

Tab. 1 Development of selected production and trade indicators of soft sawnwood and selected variables for creating models in the Slovak Republic in the years 1990-2020.

Selected indicators of soft sawnwood in the Slovak Republic in the period 1990-2020

Year Production Import Export Consumption Absolute prices

m³ m³ m³ m³ €/m3 1990a 879000 26641 165495 740146 16 1991b 641000 12200 150864 502336 26 1992b 336000 26400 165921 196479 35 1993 345000 2573 73226 274347 227 1994 400000 9600 300000 109600 132 1995 427000 11000 270625 167375 116 1996 426000 10400 241250 195150 161 1997 501000 16500 260700 256800 136 1998 845000 22600 734800 132800 389 1999 845000 18000 681000 182000 367 2000 845000 32000 683000 194000 628 2001 845000 40000 751000 134000 799 2002 845000 34000 649000 230000 605 2003 1150000 36000 645000 541000 457 2004 1251000 24000 663000 612000 581 2005 1984000 23000 681000 1326000 360 2006 1760000 56000 608634 1207366 266 2007 1872000 218000 536268 1553733 182 2008 2062861 131709 391535 1803035 124 2009 1605395 183854 354320 1434929 194 2010 1778780 235998 537005 1477773 218 2011 1460000 143066 511723 1091343 179 2012 1110000 149605 486441 773164 247 2013 990000 204926 501936 692990 233 2014 1190000 295870 695680 790190 295 2015 1150000 358000 629164 878836 441

2016 1200000 302101 810377 691724 501 2017 1305500 234517 667182 872835 388 2018 1300000 302690 794408 808282 354 2019 1263000 292392 847443 707949 447 2020 1182000 352490 958898 575592 518 Source: Faostat 2022, own calculations a – Tunák (1995) b – data calculated from date for Czechoslovakia

The values of the correlation coefficients between the explanatory and dependent variables are given in Tab. 2. Graphs of the correlation between the explanatory and dependent variables are shown in Fig. 1-6.

Tab. 2 Values of the correlation coefficient between the explanatory and dependent variables.

Dependent variables Consumption of soft sawnwood

Population 0.38

GDP per capita 0.58

GDP 0.58

Number of completed dwellings 0.52

Value of construction output 0.31

Price of soft sawnwood -0.22

Fig. 1 describes the relationships between soft sawnwood consumption and population, where a positive relationship can be observed. The results confirm the theoretical assumption that population has a positive impact on the consumption of soft sawnwood. The consumption of soft sawnwood is correlated with population (r = 0.38). Such dependencies are also pointed out by O'NEILL et al. (2010), HETEMÄKI (2011), BAUDIN (2003), BUONGIORNO (2009).

Fig. 1 Relationship between soft sawnwood consumption and population.

Fig. 2 and Fig. 3 describe the relationships between soft sawnwood consumption and GDP per capita (Fig. 2) and GDP (Fig. 3), where a positive relationship can be observed. The theoretical assumption that economic growth has an impact on the consumption of soft sawnwood can be therefore confirmed. PALUŠ (2002), HURMEKOSKI (2015) and BORZIKOWKSI (2017) also point to such conclusions. The correlation coefficient for GDP and GDP per capita has the same value (0.58). There is a presumption that a multicollinearity

will arise when classifying variables into GDP and GDP per capita models. Such an issue is solved by choosing an indicator that has better statistical parameters for modelling purposes.

Fig. 2 Relationship between soft sawnwood consumption and GDP per capita.

Fig. 3 Relationship between soft sawnwood consumption and GDP.

Fig. 4 describes the relationship between the consumption of soft sawnwood and the number of completed dwellings, where a positive relationship can be observed. Fig. 5 describes the relationships between the consumption of soft sawnwood and the value of construction output. The results confirm the theoretical assumption that construction has a positive impact on the consumption of soft sawnwood. The consumption of soft sawnwood is correlated with the activity of consumer industries – the value of construction output (r = 0.31) and the number of completed dwellings (r = 0.52). Such dependencies are also pointed out by O'CONNOR et al. (2004), ESALA et al. (2012), BAUDIN and KANGAS (2003), BUONGIORNO (1978), LUNDMARK (2010). Construction output and the number of completed dwellings are likely to be correlated when included in the models. In the USA, instead of the number of completed dwellings, the number of dwellings started is used, due to the significantly greater preference for the construction of wooden houses, which makes the consumption of soft sawnwood even higher. For example, number of completed dwellings are used by ADAMS and BLACKWELL (1973).

Fig. 4 The relationship between the consumption of soft sawnwood and the number of completed dwellings.

Fig. 5 Relationship between soft sawnwood consumption and construction output value.

Fig. 6 describes the relationships between the consumption of soft sawnwood and its price, which have a negative effect. It confirms the theoretical assumption that the price has a negative effect on the consumption of soft sawnwood. EASTIN et al. (2001), MUTANEN (2006), SATHRE and O'CONNOR (2010) point to such conclusions as well.

Fig. 6 Relationship between soft sawnwood consumption and soft sawnwood price.

The variables were selected based on the available literature. The theoretical assumption of impact was confirmed for all variables examined. The correlation coefficient between soft sawnwood consumption and its prices is r = - 0.22. The most significant 155

strength of dependence on the consumption of soft sawnwood can be observed between the change in GDP (r = 0.58) and the change in GDP per capita (r = 0.58). The signs of the values of the correlation coefficients meet the assumptions defined by the theory, i.e., that with the growth of GDP and GDP per capita, the consumption of soft sawnwood will increase (BAUDIN and KANGAS 2003, BUONGIORNO 1978). The variable of price of a wood product was represented by the variable of export price, expressed at constant 2015 prices in EUR.m-3, given that the export of soft sawnwood in the Slovak Republic exceeds the import of soft sawnwood.

CONCLUSIONS

The issue of theoretical aspects of demand models and the basis of their creation was analysed in the paper. Based on the theoretical background, the main factors relevant for the demand for soft sawnwood and their possible impact on the development of demand were defined. Different approaches to econometric modelling and their application in previous research are presented using an overview of published domestic and foreign resources. A separate part of the theoretical aspect is the analysis of wood and wood products market conditions. The main benefit of the analysis is the provision of information and the basis for demand modelling.

For modelling purposes, demand is expressed as the domestic consumption of a given product, which is calculated on the basis of the volume of production and foreign trade in a given year. For soft sawnwood, the basic explanatory variables that can be considered when creating demand models are population, GDP, GDP per capita, number of completed dwellings, value of construction output and the absolute prices of soft sawnwood. Based on these results, the factors which variables may be significant in the models of demand for soft sawnwood were determined.

REFERENCES

ADAMS, F. G., BLACKWELL, J. 1973. An econometric model of the United States forest products industry. In Forest Science 19(2), p. 82-96. ANYIRO, C.O., EZEH, C.I., OSONDU, C.K., NDUKA, G.A. 2013. Economic analysis of household energy use: A rural urban case study of Abia State, Nigeria. In Journal of Agriculture and Allied Sciences. BAUDIN, A. andBROOKS, D. 1995. Projections of forest products demand, supply and trade. In ETTS V. UNECE/ FAO Timber and Forest Discussion Papers, ETTS V Working Paper, ECE/TIM/DP/6. BAUDIN, A., BROOKS, D. 1995. Projections of forest products demand, supply and trade. In ETTS V. UN-ECE/FAO. Timber and Forest Discussion Papers. p. 41. BAUDIN, A., KANGAS, K. 2003. Modelling and projections of forest products demand, supply and trade. In Geneva Timber and Forest Discussion Paper. BORZYKOWSKI, N. 2017. The Swiss market for construction wood: estimating elasticities with time series simultaneous equations. Crag. BRÄNNLUND, R. 1989. The social loss from imperfect competition: the case of the Swedish pulpwood market. In: The Scandinavian journal of economics. - Oxford: Wiley, ISSN 0347-0520, ZDB-ID 8170-X. - Vol. 91.1989, 4, p. 689-704. BROOKS, D., SCHWARZBAUNER, P., BAUDIN, A. 1995. Modelling Forest Products Demand, Supply and Trade. United Nations, Geneva, Switzerland. p. 36. BUONGIORNO, J. 1977. Long-term forecasting of major forest products consumption in developed and developing economies. In Forest Science 23. p. 13-25. BUONGIORNO, J. 1978. Income and price elasticities in the world demand for paper and paperboard. In Forest Science. 24(2). p. 231-246.

BUONGIORNO, J. 2009. International trends in forest products consumption: is there convergence? In Int. For. Rev. 11 (4), p. 490–500. CEI-BOIS, 2004. Roadmap for the European Woodworking Industries. Executive Summary p. 34. EASTIN, I.L., SHOOK, S.R., FLEISHMAN, S.J. 2001. Material substitution in the US residential construction industry, 1994 versus 1998. In Prod. J. 51 (9), p. 30–37. ESALA, L.,HIETALA, J.,HUOVARI, J. 2012. Economic impacts of wood construction. PTT Reports 239 Helsinki, Finland. FAOSTAT. 2022. https://www.fao.org/faostat/en/#data GOLDBERGER, A. S. 1964. Econometric Theory. New York: John Wiley. GOLDSTEIN, M., M. S. KHAN. 1985. Income and Price Effects in Foreign Trade. In Handbook of International Economics. p. 1041-1105. HATRÁK, M. 2007. Ekonometria. Bratislava: IURA EDITION, 2007, Ekonómia. ISBN: 978-808078-150-7. p. 503. HENDL, J. 2004. Přehled statistických metod zpracování dat: analýza a metaanalýza dat. Praha: Portál. ISBN: 8071788201. p. 583. HETEMÄKI, L., NIINISTÖ, S., SEPPÄLÄ, R., UUSIVUORI, J. 2011. Murroksen jälkeen — metsien käytön tulevaisuus Suomessa (After the Transformation - The Future of Forest Utilization in Finland). Metsäkustannus, Hämeenlinna, p. 140. HOLMAN, R. 1999. Ekonomie. Praha. ISBN: 978-80-7400-006-5. p. 691. HURMEKOSKI, E. 2015. Factors affecting sawnwood consumption. In Europe Forest Policy and Economics, Elsevier, vol. 50, p. 236–248. HURMEKOSKI, E., HATEMÄKI, L. 2013. Studying the future of the forest sector: Review and implications for long-term outlook studies. IHEKE, O. R., OSUJI, J. 2015. Demand for fuel wood and its substitution possibilities in urban areas of Umuahia metropolis of Abia State, Nigeria. LISÝ ET AL. 2011. Ekonómia. Bratislava: Iura Edition. ISBN 978-80-8078-406-5. p. 631. LUNDMARK, R. 2010. European trade in forest products and fuels. In J. For. Econ. 16 (3), p. 235–251. MAHAPATRA, K.,GUSTAVSSON, L. 2008. Multi-storey timber buildings: breaking industry path dependency. In Build. Res. Inf. 36 (6), p. 638–648. MPRV SR. 2020. Správa o lesnom hospodárstve v Slovenskej Republike za rok 2019. Bratislava, MPRV SR, p. 70. https://www.mpsr.sk/zelena-sprava-2020/123---16162/ MICHINAKA, T., TACHIBANA, S., TURNER, J. A. 2011. Estimating price and income elasticities of demand for forest products: Cluster analysis used as a tool in grouping. In Forest Policy and Economics 13. p. 435-445. Ministerstvo pôdohospodárstva a rozvoja vidieka Slovenskej Republiky. 2022. https://www.mpsr.sk/lesy-drevo-polovnictvo/47-37-1 MUTANEN, A., 2006. Estimating substitution in coniferous sawnwood imports into Germany. In J. For. Econ. 12 (1), p. 31–50. Národná banka Slovenska. 2022. https://www.nbs.sk/sk/statisticke-udaje O'CONNOR, J., KOZAK, R., GASTON, C., FELL, D. 2004. Wood use in nonresidential buildings: opportunities and barriers. For. In Prod. J. 54 (3), p. 19–28. O'NEILL, B.C., DALTON,M., FUCHS, R., JIANG, L., PACHAURI, S., ZIGOVA, K. 2010. Global demographic trends and future carbon emissions. In Proc. Natl. Acad. Sci. U. S. A. 107 (41), p. 17521–17526. ONOJA, A. O., IDOKO, O. 2012. Econometric analysis of factors influencing fuel wood demand in rural and peri urban farm houses of Kogi State. In The Journal of Sustainable Development. PALUŠ, H. 2002. Modelovanie dopytu po výrobkoch z dreva na trhu v SR. Technická univerzita vo Zvolene. Zvolen. 1. vydanie. ISBN 80-228-1153-X. p. 49. PALUŠ. H. 2013. Trh a obchod s drevom a výrobkami z dreva. Technická univerzita vo Zvolene. Zvolen. 1. vydanie. ISBN: 978-80-228-2587-0. p. 225. SATHRE, R., O'CONNOR, J. 2010. Meta-analysis of greenhouse gas displacement factors of wood product substitution. In Environ. Sci. Pol. 13 (2), p. 104–114. SMRTNÍK, J. 1996. Mikroekonómia. Zvolen. ISBN: 80-228-0554-8. p. 173.

SOLBERG, B., MOISEYEV, A. 1997. Demand and supply analyses of roundwood and forest products markets in Europe. Overview of present studies. EFI Proceedings No. 17. European Forest Institute, Finland. ŠÚES. 2022. Štatistický úrad Európskych spoločenstiev. https://ec.europa.eu/eurostat/data/database ŠÚSR. 2022. Štatistický úrad Slovenskej Republiky. http://datacube.statistics.sk/

ACKNOWLEDGEMENTS

The authors are grateful for the support of the Scientific Grant Agency of the Ministry of Education, Science, Research, and Sport of the Slovak Republic, Grant No. 1/0494/22 Comparative Advantages of the Wood Based Sector under the Growing Influence of the Green Economy Principles and Grant No. 1/0495/22 Sustainability of Value Supply Chains and its Impact on the Competitiveness of Companies in the Forest and Forest-Based and the Slovak Research and Development Agency, Grant No. APVV-20-0294 Assessment of Economic, Social and Environmental Impacts of Forest Management in Protected Areas in SR on Forestry and Related Industries. This publication is also the result of the project implementation: Progressive research of performance properties of woodbased materials and products (LignoPro), ITMS: 313011T720 (10%) supported by the Operational Programme Integrated Infrastructure (OPII) funded by the ERDF.

AUTHORS´ ADDRESSES

Ing. Marek Hlodák Ing. Alena Rokonalová Assoc. prof. Ing. Hubert Paluš PhD. Ing. Katarína Slašťanová Technical University in Zvolen T.G. Masaryka 24 960 01 Zvolen, Slovakia marek.hlodak@gmail.com rokonalova.alena@gmail.com palus@tuzvo.sk xslastanova@is.tuzvo.sk

Prof. Branko Glavonjić, PhD. University of Belgrade Faculty of Forestry Kneza Višeslava 1 11030 Belgrade, Republic of Serbia branko.glavonjic@sfb.bg.ac.rs

This article is from: