**1. Introduction**

Over one hundred years ago, entrepreneurial activity undertaken in technologically advanced sectors was considered to be a primary source of innovation and economic growth (Schumpeter 1911). Nowadays, economic growth and competitive power are ascribed to the innovativeness of the economy to an even greater degree (Gherghina et al. 2020). From a policymaking perspective, special attention is devoted to high-tech companies and tools supporting innovative activity. Anecdotal evidence implies that the high-tech sector is a crucial driver of economic development. Furthermore, the endogenous growth theory assumes that the long-run growth rate has an endogenous character, and that the human factor plays a vital role (Kopf 2007). The decision of whether to invest more in R&D or to increase public spending on education is crucial in this context. The problem is especially important for EU countries, where, over the last decade, the lower level of investments in R&D and innovativeness has created a gap as compared to the main economic partners like the U.S. or China (Gil et al. 2019). Furthermore, investments in innovativeness create a knowledge-based society, produce intellectual capital, and finally, as (Popescu 2019) suggests, become an integral part of national wealth.

According to the results of the McKinsey Global Survey of Business Executives, on the corporate level, executives believe that innovation is the most required element of growth (Carden et al. 2005, p. 25). (Hay and Kamshad 1994), at the beginning of the 1990s, designed and performed a questionnaire based on Small and medium—sized enterprises located in the U.K. The results of the study imply that investment in product innovation was, at that time, perceived as the single most crucial strategy, followed by the policy of broadening the product range and geographic expansion.

In recent years small, medium, and young companies active in high tech sectors have attracted special attention in economic literature, as they are deemed to be a major source of innovation and development for the economy. Some authors claim that these firms have a specific business model. (Giraudo et al. 2019), Aghion and Howitt (2005), Hall (2002) stress that these firms are characterized by a specific attitude toward grasping technological innovation. Still, they also suffer from inefficient mechanisms of capital allocations, which are very severe, especially for young firms which lack track record, stable cash flows, and collaterals. (Giraudo et al. 2019) indicate that financial constraints can be especially severe for so-called bank-based economies, like Europe. Howell (2016), who investigates barriers in financing innovative firms in China, stresses that the problem of financial constraints for innovative firms can be especially severe in transitioning economies with a less developed system of financial intermediaries.

From the policymaking point of view, special attention is devoted to so-called New Technology-Based Firms (NTBFs). The term was supposedly coined by Arthur D. Little (Little 1977), who defined NTBFs as an independent venture less than 25 years old that supplies a product or service based on the exploitation of an invention or technological innovation. The issue invoked by many researchers is financial constraints, which are encountered by NTBFs at the early stage of development. So far, most studies have been focused on developed countries like the US, UK, Germany, France, or Ireland, where the institutional market environment is well established and at the same time most developed in terms of technology and science. These countries also represent a long history and have extensive experience in supporting the development of innovative activity. The high-technology firms in these countries have access to the best research centers, the best universities, and are subject to a very competitive market, and therefore their activity is based mostly on internally generated innovation.

However, scant research is devoted to the other emerging or less developed countries whose economies are trying to catch up with the leading innovators. This is especially apparent, as (Vintilă et al. 2017, p. 38) note, for countries from Eastern Europe, which endeavor to line up with Western Europe. The specific NTBFs located in these countries have other distinctive attributes. Firstly, since they are usually in emerging economies, there is no equivalent to the best research centers and access to the best universities. Secondly, there are almost no headquarters and/or research centers of multinational companies, which are usually located in the most developed countries. Thirdly, it is much more difficult for high-tech companies to compete for leading researchers with multinational companies. Fourthly, it is much more challenging to compete with high-tech companies from leading countries due to scarce resources in terms of finance, marketing, patent protection, etc.

As a result, high-tech companies in developing countries often adopt a different strategy in which innovative activity is based in substantial part on the acquisition of external technology and to a minor degree on internally generated innovation. The purchase and implementation of new technology is the preferred and less risky strategy in comparison to the development of in-house produced innovative processes. Therefore, the specificity of the high-tech companies in emerging and developing markets is slightly different in comparison to NTBFs from leading countries. The problem is especially visible within the EU, where the concept of "Two-Speed Europe" is apparent in the economic press. Therefore, as (Vintilă et al. 2018, p. 571) point out, the disparities between the West and East require a deeper understanding of proper public policy.

The main aim of the paper is the identification of determinants of the capital structure of NTBFs in a country that has an emerging economy. The focus of this study is on technological firms (NTBFs) headquartered in Poland, which is a very unique and specific case. Poland was the first CEE economy promoted by FTSE Russel's index provider with the Emerging Market to Developed Market status. Since 2018, Poland has been classified as one of the 25 most advanced global economies, including the U.S., U.K., Germany, France, Japan, etc. Therefore, Poland is considered a success story in terms of economic development, but at the same time is a country with one of the lowest levels of corporate R&D expenditures within EU countries. This contradiction urges us to investigate deeper the determinants of the capital structure of Polish NTBFs with special attention given to investments in innovativeness. We select companies at a certain stage of development that are listed on the stock exchange, mostly because of the higher quality of accounting information reported in the financial statements as compared to the non-listed companies. We hypothesize that investment in innovation has an inconclusive influence on financial leverage. Therefore, we separated it into two categories: innovation generated internally (R&D projects) and innovation acquired externally. These two types of investments have significant and distinct attributes, which we posit have a differential impact on financial leverage. We provide empirical evidence that the former kind of investment has negative, while the latter one a positive impact on financial leverage. The other hypotheses conjecture the impact of the other firm's attributes like a firm's size, liquidity, intangibility, age, profitability, and growth opportunities.

As far as we know, there is no study related to emerging economies in which investments in innovativeness are separated into externally acquired and internally generated and treated as a potential determinant of capital structure. Our hypotheses are tested on a sample of 102 firm-year observations (34 companies). The study period (2014–2018) ends at the moment when Poland was promoted to a group of countries with Developed Market status, so it can be regarded as a study of a country with the Emerging Market status.

The first section presents a literature review of the most important studies related to the problem of the financial structure of high-tech companies, the theories, and hypothesis development. The second section presents the sample characteristics, research design, and empirical results. The last section concludes with the most important issues resulting from empirical research.

### **2. Theories of Capital Structure**

Over the past several decades a number of capital structure theories have been developed which attempt to explain the creation of structures of economic entities' financing. The classical capital structure theories include *Net Income Theory, Net Operating Income Theory*, and *Traditional Theory*. Net Income Theory is based on the assumption that a firm's value is proportionate to its share of debt in capital structure, so a firm's maximum value is reached in the situation of its maximum indebtedness. Net Operating Income Theory assumes the dependence of a firm's value on the value of operating income; in the situation of determined conditions capital structure does not affect a firm's value. According to Traditional Theory, a proper balance should be maintained between internal and external sources of financing. Therefore, a reasonable level of debt increases a firm's value (Durand 1952). However, the best known classical theories are those created by F. Modigliani and M. Miller (MM). In their famous paper MM argue that a firm's value is not dependent on the capital structure but rather owners' expectations with regard to cash flows (Modigliani and Miller 1958). The conclusions based on the assumption of perfect capital markets were partly rejected in MM's next work, which took into account the issue of taxation (Modigliani and Miller 1963). MM finally admitted that indebtedness has a positive impact on a firm's value thanks to possible tax burden reductions. In the context of determining the capital structures of high-tech companies, MM and the remaining classical theories are of limited practical application (Coleman and Robb 2012; Ullah et al. 2010). High-tech firms, due to high-risk levels, do not heavily rely on debt financing; however, high debt levels have a negative impact on the value of high-tech companies.

The capital structure of high-tech firms can be more affected by the agency costs theory. Its basics were developed by Fama and Miller (1972), and initially by Jensen and Meckling (1976). It assumes the existence of conflicts of interest between owners, lenders, and managers. Managers do not always act with the intention of protecting owners' interests—they often pursue their own interests, which can be different (Novaes and Zingales 1995). In such a situation owners are forced to exercise additional control over the management's activities, which generates additional costs—agency costs. One of the ways of linking the interests of the above groups is relating the management's compensation to the company's shares. Another situation can occur in which managers implement risky investment projects, generating additional risk for lenders, while only owners benefit from higher profits. Debt can then act as a factor that disciplines the management, enforcing more active operating policies, and more effective investment policies (Kenourgios et al. 2019). In this situation, debt decreases agency costs (Novaes and Zingales 1995). Agency costs tend to be very high in companies with high unique value (Colombo et al. 2014; Sau 2007). The higher the agency costs, the lower the firm's value (Lins 2003).

A possibly significant role in high-tech firms is played by trade-off theory. Its creators are Kraus and Litzenberger (1973). All financing methods have both advantages and drawbacks. Higher debt levels provide an opportunity to deduct interest from taxable income. However, it should be noted that there are other methods of reducing tax burden with the use of non-interest tax shields including effective depreciation policies, or, in a broader sense, tax optimization (DeAngelo and Masulis 1980). It should be stressed that a company can benefit considerably from relatively high tax rates. A company's heavy reliance on indebtedness in its capital structure increases business risk and results in the costs of bankruptcy (Baxter 1967). The higher the bankruptcy costs, the lower a firm's value. Higher debt levels in the balance sheet total originally increase a firm's value, but at a certain point, a firm's value decreases (Adrienn 2014). The costs of bankruptcy are then higher than tax shield positive effects. A practical confirmation of the trade-off theory is the occurrence of the so-called industry effect. The functioning of an enterprise within one industry is dependent on similar factors—economic entities are characterized by similar operating cycles, risk levels, and agency costs, hence their similar share of debt in overall financing. The companies whose share of debt in the structure of financing is below industry average tend to increase it, unlike entities that have a large share of debt in their financing structure and try to lower its level (K ˛edzior 2012). The industry effect is not identical in all industries. In industries characterized by stiff competition and diversified agency costs, debt levels can vary. Unequal access to advanced technologies has a similar impact on indebtedness (Michaelas et al. 1999). The above factors result in the existence of an optimal industry capital structure, which economic entities seek to achieve in their long-term operations (M'ng et al. 2017).

In the case of innovative companies, it is difficult to estimate the risk of the sources of financing within the framework of trade-off theory. Many threats should be regarded as potential, and their materialization is conditional and not easy to estimate (Sau 2007), hence difficulties in choosing adequate sources of financing. Choices made by high-tech firms with regard to financing are affected by a rapidly changing business environment and the complexity of applied technologies (Li et al. 2006). These entities do not have the ability to offer adequate guarantees to mitigate lenders' risk (Sau 2007). Innovative firms have higher bankruptcy costs (Aghion et al. 2004; Sau 2007), so the share of liabilities in the balance sheet total cannot be dominant. High-tech companies with a relatively high volume of intangible assets are less inclined to borrow funds. On the other hand, high growth companies rely on debt financing to a smaller degree (Castro et al. 2015). Transaction costs in such entities are also high due to risk factors and, generally, limited volumes (Revest and Sapio 2012). Their market value is subject to large fluctuations, especially as their financial standing deteriorates. It results from the fact that their valuation is based on specialized assets as well as large growth potential. Therefore, valuation changes on stock exchanges play a crucial role in high-tech firms (Revest and Sapio 2012).

The financial conditions and capital requirements of high-tech firms depend on the stage of their development (Sau 2007). At the initial stage of development economic entities' cash flows are often negative, so they are not able to repay their debts, and the acquisition of funds is difficult. In their early stages, high-tech firms' biggest problem in product commercialization based on the use of familiar technologies is the acquisition of funds for operating activities (Minola et al. 2013).

The creation of capital structure is greatly affected by the pecking order theory. The theory was created by Donaldson (1961), and then elaborated and modified by Myers and Majluf (1984). The authors divide sources of financing into external and internal sources. The choice of the sources of financing is mainly determined by their cost which is lower for internal capital. Therefore, companies should finance their operations by relying on retained earnings, followed by debt and, finally, the issue of shares (Stulz 1990). This order is justified by information asymmetry in relationships between companies, banks and external investors. Banks and external investors have more difficulty accessing information about companies than people operating within company structures, so in light of the higher risk of transferring capital, they require higher interest on loans and higher rates of return. Information asymmetry leads to moral hazard and adverse selection. The adverse selection indicates that banks find it difficult to distinguish between effective and ineffective investment projects, which generates additional costs and increases risk. A high level of adverse selection also results from great uncertainty with regard to future return on investment rates as compared with traditionally implemented projects (Carpenter and Petersen 2002). Moral hazard indicates that owners benefit more from implementing risky investment projects than debtors (Aoun and Heshmati 2006). High information asymmetry results, to a considerable degree, from the large development potential of high-tech firms (Castro et al. 2015). High information asymmetry in the technology sector mainly applies to small companies. Therefore, such companies can often be undervalued (Coleman and Robb 2012).

Pecking order theory assumes that the accessibility of information about a high-tech firm has an impact on the choice of capital structure. To avoid problems resulting from the disclosure of internal information to a larger group of stakeholders, high-tech firms give preference to internal sources of financing (Hogan et al. 2017; Scherr et al. 1993). Due to such factors as uncertainty with regard to the ultimate results of innovative investment projects, possible cases of underinvesting and overtrading, difficulties in monitoring R&D activities, and the frequent lack of comprehensive knowledge about technology among investors and banks, access to external financing can be limited (Revest and Sapio 2012). Generally, high reinvestment rates in technology firms force them to seek external sources of financing in the absence of their own funds (Berggren et al. 2000).

The acquisition of external capital implies the necessity of disclosing additional information about planned operating or investment activities. Small and medium-sized high-tech firms are not inclined to disclose such information. Similar opinions are held by Revest and Sapio (2012). Technology firms are unwilling to disclose detailed information about R&D programs due to a very competitive market and the fear of losing competitive advantage. Aoun and Heshmati (2006) also claim that because of the confidential character of business operations high-tech firms have difficulty disclosing comprehensive financial data, and hence face problems with acquiring funds for business activities. As a result, markets do not possess full information, and lenders have limited knowledge about the current operations of high-tech firms (Ullah et al. 2010). Transaction costs and greater flexibility of operations justify reliance on retained earnings as a source of financing (Grinblatt and Titman 2002). A number of empirical research studies point to a negative correlation between profitability and indebtedness (Bhayani 2010a; Korkmaz and Karaca 2014). Therefore, profitable firms rely on debt financing on a limited scale.

Technology firms tend to choose financing through the issue of shares rather than indebtedness. This mainly refers to young firms at an early stage of development (Minola et al. 2013). Innovative firms are characterized by attractive investment possibilities as compared with other business entities, but the costs of the issue of shares should be regarded as high (Aghion et al. 2004; Castro et al. 2015). Larger technology firms have a greater ability to raise funds through the issue of shares (Mac an Bhaird and Lucey 2010). Frequently, young firms without a long credit history and relationships with banks are left with no other option but to issue shares (Carpenter and Petersen 2002). Because of the lack of collateral in the form of tangible assets, innovative companies tend to rely more frequently on share capital. The issue of shares does not have to be secured by tangible assets and does not increase the threat of bankruptcy. High-tech firms can successfully implement R&D programmes if they are able to convince investors to purchase issued shares (Carpenter and Petersen 2002). The idea

of the issue of shares is also justified by technology firms' tendency to implement high returns but also risky investments (Carpenter and Petersen 2002).

Because of the risk of share dilution and takeovers, innovative companies tend to rely on debt financing (Aghion et al. 2004). If the lack of transparency of disclosures is acceptable, high-tech firms can also resort to bank loans (Berggren et al. 2000). As firms grow and gain more experience, the range of information asymmetry reduces, the value of assets (especially tangible assets) increases, and access to bank loans becomes easier (Hogan et al. 2017). High-risk firms may not be granted loans, but they are still able to successfully implement the process of issuing shares.

It seems, however, that pecking order theory turns out to be more useful in large economic entities, which rarely issue shares because of the high values of retained earnings and the possibility of acquiring corporate bonds (Akgül and Sigali 2018). Nevertheless, within a short time horizon, enterprises are likely to create their capital structure based on the pecking order theory. On the other hand, in longer periods of time in which the changeability of cash flows and economic conditions is less severe, companies are likely to rely on trade-off theory (Bontempi 2002). Pecking order theory (POT) is probably more effective in describing the choice of sources of financing in mature companies as compared with high growth entities.

The signaling theory, created by Ross (1977), has a different impact on capital structure creation. Due to information asymmetry, people operating inside and outside of an organization have unequal access to information about a company's financial standing. External stakeholders make intensive efforts to obtain information about a company's future financial condition and future share valuations. Therefore, they seek additional signals concerning an economic entity's actual financial condition. The most reliable signals and those that cannot be easily imitated refer to dividend policies and capital structure decisions (Frankfurter and Wood 2002; Deesomsak et al. 2004). Increased indebtedness should be regarded as a positive signal—it indicates a bank's favorable assessment of an entity's creditworthiness and stable projected financial results and cash flows. It can be assumed that current and projected financial results will not be diluted. On the other hand, the issue of shares is sometimes treated by financial markets as a negative signal. Companies with less optimistic financial result predictions tend to finance their operations through the issue of shares (Leland and Pyle 1977). Managers choose the issue of shares if their current valuation is excessively high. The market's negative response to the issue of shares can be even more severe if investors perceive a company as being characterized by great information asymmetry (Minola et al. 2013). The range of information asymmetry is very high for new investment projects, new areas of activity and new strategies. Over time asymmetry tends to decrease (Harris and Raviv 1988). Information asymmetry relates to new areas and issues—past events of key significance are reflected in the price of shares (Harris and Raviv 1988).

Last but not least important is the financial life cycle theory, which assumes that a firm's capital structure preferences vary with their life cycle (Butzbach and Sarno 2018). The life cycle determined the availability of financial resources and the cost of capital. The theory implies that smaller and younger companies exhibit higher information asymmetry, which in turn increases the cost of capital. We expect that in the case of NTBFs the financial life cycle theory may be of use due to the fact that R&D activity increases information asymmetry and the fact that NTBFs are typically young companies with low or no reputation, and have almost no (or very low) carrying amount of tangible assets.
