The Impact of Culture on Capital Structure: Individualism, Innovative Culture and Leverage

The Impact of Culture on Capital Structure: Individualism, Innovative Culture and Leverage

Anyu Chen 11639547

University of Amsterdam

Business Administration – Finance

General topic: Country culture and corporate financial policies Slot: The effect of company culture on capital structure

Supervisor: Dr. Evgenia Zhivotova

Statement of Originality

This document is written by Anyu Chen (11639547) who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Abstract

There are numerous literatures finding that both corporate culture and national culture have effects on a firm’s capital structure, however, there is less discussion on the potential relationship between these three topics. Therefore, this paper assesses whether there is a potential indirect effect of individualism on capital structure that can be explained by innovative corporate culture. We analyze 24,075 firms’ leverage in 28 countries and provide evidence that there is a direct effect of individualism on a firm’s leverage instead of an indirect effect. The results suggest that firms that have a more innovative culture are at higher leverage positions, on the other hand, firms that located in individualism societies also have higher leverage. However, the relationship between individualism and innovative culture is not significant, which no indirect effect of individualism on leverage is explained by innovative culture. The paper finds that both national culture and corporate culture are important determinants of a firm’s capital structure.

INTRODUCTION

Culture embeds deeply in a surrounding that affects people’s decisions, opinions and behaviours, it is difficult to be measured, but also too influential to be ignored. There are various definitions of culture, in summary, Guiso, Sapienza and Zingales (2006) define culture as “customary beliefs and values that ethnic, religious, and social groups transmit fairly unchanged from generation to generation”. Culture is a factor that impels many differences in all fields, however, it used to be ignored by economists who argued that “it does not permit him to predict outcomes” (Harrison and Huntington, 2000). Due to the characteristics of culture, it is difficult to be measured in quantitative research. Thus, it is plausible to understand economists’ view on culture. Nevertheless, there is more information available as the data collection techniques advanced, which the comprehensive data provides alternative ways to quantify culture in financial researches. Gleason, Mathur and Mathur (2000) have proposed that culture is an influential factor on capital structure, which is concluded from retailers in 14 European countries. The research dispels the worries of economists that raised in Landes’ book, it shows that culture can be explanatory and towards a predictive outcome in financial research.

Based on the research of Gleason et al. (2000), we argue that culture is a potential predictor of capital structure, other than the apparent firm-level factors. According to the Pecking order hypothesis, the financing decision follows the order of retained earnings, debt and equity as the last resort (Myers and Majluf, 1984). However, the hypothesis is not explanatory in every case, Chen (2004) proposed a “new pecking

order” that suits to Chinese firms. The research suggests that Chinese firms also tend to choose the retained earnings as the optimal option, but they prefer equity funding over debt issuing. Chen (2004) clarifies that the discrepancy is a result of fundamental institutional assumptions difference to the West. The fundamental institutional assumption is rather broad, however, Chui, Lloyd and Kwok’s (2002) research explain it further that culture is the hidden factor that determines capital structure.

The majority research of cultural influence on capital structure is at the firm-specific level, there is relatively less attention to country culture. Firstly, this study focuses on the preceding researches on corporate culture and capital structure. Then, the relationship between country culture and corporate culture will be investigated to discover whether there are potential effects between country culture, corporate culture and capital structure. O’Brien (2003) has concluded that innovation is a significant indicator of a company’s leverage level. In O’Brien’s research, the empirical result is consistent with the previous studies on organizational slack. The continuous innovation activities require stable support of fund, a financial slack is desired for successful innovation investment. Financial slack is the amount of cash and marketable securities on hand, which saves time and troubles to better capture innovative investment opportunities than debt financing. In addition, the outcomes of innovation investment are mainly classified as intangible assets, such as patents and human resources, which are ineligible as collateral (Long and Malitz, 1985, as cited in O’Brien, 2003). It makes debt financing more costly and higher risk of bankruptcy when firms default, therefore, firms will not borrow as much when financial slack is more efficient for funding

innovative projects. Hence, O’Brien (2003) has observed a negative correlation between a firm’s innovativeness and its leverage. Regardless of the effect’s direction, all researches advocate that innovation is an important determinant of capital structure. In this paper, the interpretation of innovation is distinct from O’Brien’s (2003) research, which we treat innovation as a corporate culture instead of a specific strategy. Sathe (1983) defines corporate culture as “the set of important understandings that members of a community share in common”. These important understandings are essentially the values and beliefs shared by corporate’s members, which can be reflected by observable behaviours (Schwartz and Davis, 1981). For instance, R&D expenses is an observable behaviour that concerns the strategic aspect of innovativeness, however, single observable behaviour is too narrow to represent the common understandings that shared within the corporation. Therefore, based on the previous researches on innovative culture, we measure innovativeness from multiple dimensions that are market orientation, strategic orientation and process innovation (Dobni, 2008; Wang and Ahmed, 2004). We believe an effective implementation of strategy should be coherent with its corporate culture, which in this case, an innovation strategy is essentially designed and pursued under an innovative culture (Naranjo-Valencia, Jiménez-Jiménez and Sanz-Valle, 2011; Shrivastava, 1985). On the other hand, the relationship between innovativeness and leverage may be different from O’Brien’s findings. O’Brien (2003) has assumed that intangible assets cannot be treated as collateral. However, there has been an increasing trend of collagenization of intangible assets, which implicates a possible invalidation of O’Brien’s assumption.

On the national cultural level, the previous researches are primarily focusing on the Western countries, for instance, Gleason’s et al. (2000) data is about 14 European countries and O’Brien (2003) is more focus on the companies in the United States. Oyserman, Coon and Kemmelmeier (2002) observe that Chinese are more collectivistic than Europeans, Africans and Americans. The difference could lead to divergent decision making and organizational structure as they put different weight on individual and group. The self-concept and cognition disparities have an impact on institutional logics (Thornton, Ocasio and Lounsbury, 2015). It could be a fundamental institutional difference as Chen (2004) proposed, which leads to different preferences on capital structure. Moreover, high individualism is wildly recognized as the determinant of higher innovation. In an individualism society, people have more independence, autonomy and freedom than a collectivism society, which is helpful to facilitate innovative activities (Shane, 1993). The relationships between individualism, innovativeness and capital structure also imply that national culture, such as individualism/collectivism, is also a potential factor that determines a firm’s leverage.

The country culture and corporate culture are the two main focuses of this paper. The goal is to investigate how individualism affects firms’ innovative culture and their capital structure, whether individualism has direct effects or indirect effects on leverage that is mediated by innovative culture.

Analysis of capital structure

One of the earliest influential theory on the capital structure was introduced by Franco Modigliani and Merton Miller. The M&M propositions I and II elaborate that, in a perfect market, capital structure is irrelevant to a firm’s value and the cost of capital (Modigliani and Miller, 1958). Based on the theorem, firms should not be bothered by capital structure since it has no effect on its value and performance. Nevertheless, all of the M&M propositions are developed under assumptions of a perfect market and a world without taxes, which neither is true in real-world applications. If taxes are taken into consideration, the deductibility of interest payment on taxes increases firms’ value, which generates an incentive for firms to increase leverage as much as possible (Modigliani and Miller, 1963). Therefore, the early version of M&M theorem does not hold any longer since the value of firm changes as the decision to leverage diverges. On the other hand, managers have the access to the first-hand firm-specific information before the investor does. Consequently, a perfect market is never the case in real-world applications where information asymmetry is the major obstacle. Although the asymmetry information can be alleviated through information release (e.g. equity issue announcements and earning announcements), it can hardly be eliminated (Dierkens, 1991). Therefore, M&M theorem is persuasive in an ideal environment, more explanations are required for real-world applications.

Kraus and Litzenberger (1973) have provided more explanation on capital structure towards to the Modigliani and Miller’s (1963) tax correction, which he has proposed that there is a trade-off on determining leverage ratio. Although firms benefit from the

tax deduction on interest payment, the growing debt exposures the firm to a higher probability of bankruptcy. The bankruptcy penalty is another factor that challenges the perfect capital market assumption, the higher cost of penalty the less incentive of debt financing. Therefore, Kraus and Litzenberger (1973) suggest that, in consideration of the trade-off between taxes and bankruptcy, there is an optimal leverage ratio for each firm and the firm’s performance and value are not independent of its capital structure.

In contrast to Modigliani and Miller (1958), Myers and Majluf (1984) presume that management possesses more information than the market, which the asymmetry of information affects a firm’s choice on capital structure. The preference is known as pecking order theory. The asymmetric information causes agency issues and different cost of financing. If management knows more information than the market, they likely to exploit this advantage to place their interest above the new investors’. It is problematic during equity issuing because it risks the firm’s opportunity to obtain a positive NPV project. Myers and Majluf (1984) then suggest that default-risk-free debt or even risky debt can mitigate the conflict between the old and the new shareholder, especially when internal financing is not available. The research not only emphasized the importance of financial slack for reducing agency issue but also in support of the idea that capital structure matters.

What is innovative culture

One previous research defines organizational innovativeness as:

market, or opening up new markets, through combining strategic orientation with innovative behavior and process (Wang and Ahmed, 2004).

Figure I: model of innovation culture

Innovation has been extensively investigated by researchers, especially after Porter’s analysis of the competitive market. Porter (1986) has suggested that both the domestic and global market have been becoming more competitive as globalization popularizes, which stimulates higher participation in international trade and foreign direct investment. Therefore, Porter suggests differentiation is one of the core competitive advantages that enable firms to survive in competitions, in both services and products. As the engine of differentiation, innovation has been found with influential effects on corporation performance (Jansen, Van Den Bosch and Volberda, 2006; Verhees and Meulenberg, 2004). Fundamentally, innovation helps firms to achieve better performance by differentiating their products and services from their rivals to diminish the competition.

Wang and Ahmed (2004) advocate that organizational innovativeness cannot only be measured uni-dimensional, as either a strategy or behaviour. The research approaches innovation measurement from five levels that are product innovativeness, market innovativeness, process innovativeness, behavioural innovativeness and

strategic innovativeness. The development of new products is only the outcome of innovation, however, the underlying elements that achieve the outcomes are innovativeness of a corporation. Dobni’s (2008) research is consistent with Wang and Ahmed (2004) that it is more reliable to observe innovation culture from multi-dimensional level, which focuses on four dimensions that are innovation intention, innovation infrastructure, market orientation and innovation implementation. Although these studies prescribe different elements on measuring organizational innovativeness, there are much in common of the underlying concepts. Therefore, based on both research, a model of measuring innovation culture is constructed, as in Figure I.

Innovation culture and capital structure

In supplement of the model in Figure I, precedent studies provide empirical evidence that links between each of the dimension and debt financing. Olavarrieta and Friedmann (1999) have emphasized the importance of innovation based on Porter’s (1986) competitive advantage. The knowledge-related resources and hard-to-imitate resources, such as patent and skills, which are essential for a firm in an exceedingly competitive environment. In other words, these intangible assets reflect on the firm’s efforts on market orientation for innovation to remain competitive. Most of the previous researches have suggested that firms with high proportion intangible assets tend to have lower debt ratio, as result of difficult valuation on assets and effectiveness of bond covenants (Long and Malitz, 1985; O’Brien, 2003). Nevertheless, these researches underestimate the value of intangible assets and their eligibility as collaterals. Lim et

al. (2020) have pointed out that firms also have high leverage, even possess a high proportion of intangible assets. There is a slight difference in the form of debt, which convertible debt, term loan and bank loan are the primary sources. Firms are endowed with the ability to borrow as the higher potential value and transferability of intangible assets. Therefore, we propose a hypothesis that:

Hypothesis 1: Firms with more intangible assets, the higher its leverage will be. On the other hand, in the dimension of process innovation, scholars have generally recognized that the greater effort into innovation, the lower of the production costs will be (Mitchell and Coles, 2003; Powell, Fu, Horowitz, Basore, Woodhouse and Buonassisi, 2015). Powell et al. (2015) also have associated it further with leverage ratio in the photovoltaics industry, which the result is consistent with the Trade-off Theory. The increasing leverage ratio is beneficial to the innovation as well as business process until an optimal capital structure is reached. The insufficient process innovation leads to underperformance in the industry, as a result, even more, debt financing is used to catch up the pace of the industry. Therefore, insufficient process innovation lead firms to a higher leverage position:

Hypothesis 2: There is a negative correlation between the cost of goods sold and the leverage ratio.

Finally, Hambrick and Macmillan (1985) have found that R&D expenditures are an indication of innovation. Although the output of R&D investment may not efficient as expected, the intention is clear, which is to obtain new technologies, known-how (intellectual properties) and market power. As a consequence, R&D is a representation

of innovation on the strategic dimension. O’Brien (2003) has associated it further to capital structure that, a firm with relative higher R&D expenses in an industry possesses a lower debt ratio than the others. Ample financial slack ensures that firms have the ability to capture most of the innovative ideas and place them into action. Therefore, a firm value more on innovation, the less leverage and more retained earnings kept for the innovation process. Similarly, Czarnitzki and Kraft (2009) argue that there is a negative correlation between innovativeness and debt financing when the interest payment is a potential obstacle to R&D investment or even the cause of bankruptcy. Nonetheless, Czarnitzki and Kraft (2009) also have doubt on financial slack regarding agency problems. The existence of agency problems offset the advantage of financial slack on R&D investment, such as squandering of the fund and inefficient investment (Rajan, Servaes and Zingales, 2000; Denis, Denis and Sarin, 1997). In fact, in order to meet the interest payment and avoiding bankruptcy, debt disciplines the management to handle the capital properly (Czarnitzki and Kraft, 2009). Furthermore, default-risk-free debt is also a source of financial slack (Myers and Majluf, 1984), thus, there is a possibility of positive effect between innovativeness and leverage that in the condition of appropriate debt ratio.

Admittedly, internal funding is not available all time when an investment opportunity occurs, especially for a firm that has low operating profitability. Bartoloni (2011) has observed that profitable firms tend to follow the pecking order theory that is in favour of using internal finance. However, when internal finance is not sufficient for meeting innovative investment, all firms gravitate towards debt financing as

innovativeness increases. Although the result contradicted O’Brien’s finding, it seems can be explained by Myers and Majluf (1984)’s definition of financial slack:

The sum of cash on hand and marketable securities will be referred to as financial slack…Financial slack should also include the amount of default-risk-free debt the firm can issue (Myers and Majluf, 1984).

Debt financing is a source of financial slack as long as firms are able to fulfil their debt obligation, which then is beneficial for innovative activities. Based on these arguments, we propose a hypothesis as follows:

Hypothesis 3: Firms that have higher R&D expenditure, the higher its leverage ratio will be. There is a positive effect of innovative strategy on leverage.

Individualism, innovative culture and capital structure

Hofstede (2010) clearly differentiates the national culture and organizational culture, where national culture is more incorporated in the human mind than organizational culture, also, national culture is hard to change from generation to generation (Guiso et al., 2006). Therefore, individualism-collectivism can be classified as national culture traits as it persistently differs across geographic locations and cultural backgrounds, it is derived from a social system rather than an individual basis (Hui, 1988). Geert Hofstede was one of the earliest researchers to measure national culture and value, the framework constructed by Hofstede is as known as Hofstede’s culture dimensional theory. He defines individualism as “the emotional independence from groups, organizations or other collectivises” (Hofstede, 1980, p.221). In contrast,

collectivism values more on group decisions, in which people concerns for other’s feeling and sharing of outcomes (Hui and Triandis, 1986).

Shane (1993) has investigated individualism further and found a higher rate of innovation in individualism societies than collectivism societies. The research assumes three traits between individualism and collectivism societies, which are freedom, autonomy and independence. For instance, in a collectivism society, people have less incentive to express creative ideas in order to maintain harmony in the group. Hence, the individual’s idea is likely to be restricted by group thinking, on the other hand, individual’s independence, autonomy and freedom in individualism societies have been found to encourage creativity and innovative activities (Goncalo and Staw, 2006). The result has also been found within corporations. Chen, Podolski and Veeraraghavan (2017) have observed that firms located in individualism societies have more influential patents and more efficient R&D investments than the firms in collectivism societies. Therefore, we expect to find firms have higher innovativeness in individualism societies than collectivism societies. Based on the innovative corporate culture in Figure I, we propose hypothesises that:

Hypothesis 4: Firms located in high individualism societies have high market orientation for innovation. Firms hold more intangible assets in individualism societies. Hypothesis 5: Firms located in high individualism societies have higher R&D investment intensity than collectivism societies.

Hypothesis 6: Firms located in high individualism societies have higher process innovation than collectivism societies, which would have a lower COGs.

If a positive relationship between innovative corporate culture and leverage is observed, based on these hypothesises described above, there is a possible correlation between individualism and firm’s leverage. Antonczyk and Salzmann (2014) have discovered a positive relationship between the level of individualism and debt ratio, however, they approach this question from a different perspective. They have found that people are more overconfidence and optimism in individualism societies. Managers may overvalue the return of projects and the ability to repay debt, which generates an upwards bias when determining debt ratios. Therefore, we propose a hypothesis that:

Hypothesis 7: Firms located in individualism societies have higher leverage than collectivism societies.

METHODOLOGY

Data description

The firm-specific data used in this study were collected from Compustat Global, where contained comprehensive firm-specific financial information across countries. The data has detailed information on the firm’s balance sheet that was originally acquired from annual reports to shareholders. Regarding the national culture factor, it was the base culture data in Geert Hofstede’s research. The data included 6-dimensions of culture from 112 countries, which were power distance, uncertainty avoidance, individualism or collectivism, masculinity or femininity, long- or short-term orientation.

All of the dimensions were concluded from a cross-national survey though IBM database, which amounted to 100,000 questionnaires.

The preliminary sample included 5 years of data in between Jan 2015 to Jan 2020. There were 169,033 observations on 37,387 companies across 117 countries. The population of the study was all companies in Compustat Global, however, Compustat Global had a lack of data in some countries that was not enough for statistical analysis. Also, Hofstede’s data and Compustat Global database had some difference in countries that included. Therefore, a robustness check was implemented to exclude overrepresented observations. The original dataset had missing inputs on multiple variables. There were missing data on firm’s total assets and any observation with missing book value of total asset was excluded from the sample, since the calculation of dependent variable and independent variables were based on the book value of assets. Furthermore, the duplication test detected repeated time values because some firms reported multiple times in one year, which the duplicates had also been dropped. In addition, the dependent and independent variables in the regression analysis were lagged for 1 year, which the samples that had less than 2 years observations were not satisfied with regression analysis in all models. In the end, a total of 109,471 observations were satisfied for further analysis, which accounted for 61% of the original dataset.

Dependent Variable

is the firms’ leverage. Model 1 is used to test the hypothesises 1, 2 and 3, whereas model 3 is constructed for testing hypothesise 7. The leverage is calculated by dividing the book value of total debt by the book value of the total asset, in which total debt is the sum of total short-term and long-term debt. The debt-to-asset ratio is a good indicator that represents the firm’s insensitivity on debt financing. Model 2 is used to test the hypothesises 4, 5 and 6, whether the level of individualism has an effect on a firm’s innovativeness. Therefore, the dependent variables for model 2 are R&D expenses, amount of intangible assets and cost of goods sold, these represent measure innovativeness on strategic orientation, market orientation and process innovation respectively. There is a concern of nonlinear correlation effects between dependent variables and independent variables, therefore, we employ a non-linear test (-nlcheck-) to verify potential nonlinear correlations. As a result, average industry leverage, profitability and average industry profitability exhibit insignificant linear relationship with the dependent variable. We have tried different transformation of these variable and the natural logarithm transformation has the best model fit for all the variables.

Independent Variable

To test the first three hypothesises that mentioned in the theoretical framework, three primary independent variables affect firm’s decision on leverage, which are market orientation for innovation, strategic orientation for innovation and process innovation. First, this study perceives the orientation of innovation as the amount of knowledge-related resources, which is computed as the natural log of the total

intangible assets. Second, the strategic orientation for innovation is measured in innovation intensity, which is the R&D expenses to sales ratio. Similar logic applies, relative innovation intensity is more appropriate for analysing the difference between firms. We compute the innovation intensity by dividing R&D expenses by total sales, then relative innovation intensity is calculated by subtracting the R&D to sales ratios from the average industry’s R&D to sales ratios. Although there are firms with missing R&D information in the dataset, we assume that these firms have no R&D expenses by following the practice of the most financial researches (Singh, 2008; O’Brien, 2003). Finally, we use the relative cost of goods sold to revenue ratio to measure the firm’s efforts on process innovation. Again, the relative COGs ratio is computed by subtracting the firm’s COGs ratio from the average industry’s COGs ratio. For testing hypothesis 7, the primary goal is to evaluate whether individualism has effects on the leverage ratio, thus, the degree of individualism for each country is included. We directly obtained the data from Hofstede indices, which the individualism is originally measured by “I” consciousness, the importance of personal opinion and privacy, weighting task over relationship and other criteria (Hofstede, 2010).

In order to focus on the primary independent variables’ effects, we choose to control variables both on firm-level and industry level. In the aspect of firm-level, the variable profitability controls for the return on assets, which measured by dividing the operating income (before depreciation) by the total assets. According to the previous studies and the pecking order theory, firms that with adequate retained earnings are at lower leverage ratio, which they generally prefer fund investments with retained

earnings than issuing debt (Myers and Majluf, 1984). Therefore, we also control for the amount of retained earnings that firms have each year. The size of retained earnings is measured by taking the natural log. A common proxy for the firm-size is also taking the natural log of the book value of the total assets. Moreover, there are more controls on industry leverage and profitability, in which these are calculated from the average value of leverage and profitability in each industry. On the other hand, in order to focus on the effect of individualism in model 2 and model 3, we also control for other country culture factors that included in Hofstede culture index. For instance, power distance, masculinity, uncertainty avoidance, long-term orientation and indulgence are controlled for testing model 2 and model 3.

All variables are collected and computed annually. The Pearson correlation test shows that there are no or weak correlation between all variables described above, the details are showed in the descriptive statistics table in Table 1. Although there are few moderate correlations, most of the variables have no correlation or weakly correlation with each other.

Analysis

To examine the hypothesises that have described above, we have conducted some preliminary tests to verify the variables and the structure of regression models. Otherwise, an inappropriate choice of regression model and variables would lead to an unreliable testing result. To start with, there are two common statistical models for panel data regression that are fixed-effects model and random-effects. In general, the fixed

effect has a group variable that helps to control time and homogeneous within the specified group. To test the hypothesises 1, 2 and 3, all the variables changes over time at firms-specific level, which we should expect homogeneous observation of a firm in multiple time periods. The characteristic of the fixed-effects model eliminates omitted variable bias when there are variables omitted, as well as controlling for the unmeasured effects. Additionally, the result of the Hausman test also concludes that the fixed-effects model would be a better choice for testing hypothesises 1, 2 and 3 (Pro>chi=0.000, significant at 0.05). However, there is a concern of Nickell bias for the fixed-effects model that includes lagged independent variables, especially for small T (time) and big N (observations) (Nickell, 1981). The bias is essentially caused by the endogeneity between lagged variables and the error term. Therefore, instead of a fixed-effect, we implement Anderson-Hsiao estimator to mitigate the correlation between regressor and error. Furthermore, firms share some similarities within the same industry, there are still discrepancies between them. As a result, the heterogeneity is controlled by clustering the industries, which we assume that firms in the same shares the same variance.

On the other hand, the fixed-effects model includes both observed and unobserved group-specific effect, therefore, it has constraints on predicting variable that has no within-group effect (Ashenfelter, Levine and Zimmerman, 2003). According to the idea, it is inappropriate to use the fixed-effects model to test for hypothesises 4, 5, 6 and 7. The variable of individualism only has one value for all the firms in the same country, which has no effect within groups (company id) and it is consistent over time. Random-effects can predict outcomes at different levels, in this case, it is ideal since our interest

is on the national level that is beyond the company-specific level.

Finally, we use both lagged dependent variables and independent variables when testing hypothesises 1, 2, 3 and 7. The reason is that there are probabilities firms are consistent with their R&D expenses, COGs and amount of intangible assets, which then these variable’s effects are negligible and will not be accounted for in the model. For this reason, the lagged dependent variable (leverage) is also treated as the predictor variable, which adds explanatory power to the model (O’Brien, 2003). All the other independent variables are also lagged, it is a common practice in leverage researches (Adrian and Shin, 2010; O’Brien, 2003). According to Tsyplakov’s (2008), there is a time-to-build factor that is the time lag between the start and end of an investment project. An appropriate measure or prediction of investment projects is critical when making strategic decisions, however, the result will be different if the time-to-build factor is ignored. Consequently, there will be various leverage preferences between firms in the same industry or across industries as they have different time-to-build on investment projects. Therefore, we have lagged all the independent variable by 1 year to control for the time-to-build factor, which provides independent variables with time to reflect on leverage. In other words, the model is testing the leverage difference between time t and t - 1, whether it is affected by the independent variables at t - 1. In contrast, lagged individualism is not used for testing hypothesis 4 because the individualism is hardly changing over time.

(1) 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒 = 𝛽_!+ 𝛽_"𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒_#$"+ 𝛽_%𝑅&𝐷 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦_#$"+ 𝛽_&𝐼𝑛𝑡𝑎𝑛𝑔𝑖𝑏𝑙𝑒_#$"+ 𝛽_'𝐶𝑂𝐺𝑠 𝑟𝑎𝑡𝑖𝑜_#$"+ 𝛽₍𝑃𝑟𝑜𝑓𝑖𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦_#$"+ 𝛽₎𝑅𝑒𝑡𝑎𝑖𝑛𝑒𝑑 𝑒𝑎𝑟𝑛𝑖𝑛𝑔𝑠_#$"+ 𝛽_*𝐹𝑖𝑟𝑚 𝑠𝑖𝑧𝑒_#$"+ 𝛽₊𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝑙𝑒𝑣𝑒𝑟𝑎𝑔𝑒_#$"+ 𝛽_,𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝑝𝑟𝑜𝑓𝑖𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦_#$" (2) 𝑅&𝐷 𝑜𝑟 𝐼𝑛𝑡𝑎𝑛𝑔𝑖𝑏𝑙𝑒 𝑎𝑠𝑠𝑒𝑡𝑠 𝑜𝑟 𝐶𝑂𝐺𝑠 = 𝛽_!+ 𝛽_"𝐼𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙𝑖𝑠𝑚 + 𝛽_%𝑃𝑜𝑤𝑒𝑟 𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 + 𝛽_&𝑀𝑎𝑠𝑐𝑢𝑙𝑖𝑛𝑖𝑡𝑦 + 𝛽_'𝑈𝑛𝑐𝑒𝑟𝑡𝑎𝑖𝑛𝑡𝑦 𝑎𝑣𝑜𝑖𝑑𝑎𝑛𝑐𝑒 + 𝛽₍𝐿𝑜𝑛𝑔 𝑡𝑒𝑟𝑚 𝑜𝑟𝑖𝑒𝑛𝑡𝑎𝑡𝑖𝑜𝑛 + 𝛽₎𝐼𝑛𝑑𝑢𝑙𝑔𝑒𝑛𝑐𝑒 + 𝛽_*𝐹𝑖𝑟𝑚 𝑠𝑖𝑧𝑒 + 𝛽₊𝑃𝑟𝑜𝑓𝑖𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦 + 𝛽_,𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝑝𝑟𝑜𝑓𝑖𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦 *R&D, intangible assets and COGs are tested separately.

(3) 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒 = 𝛽_!+ 𝛽_"𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒_#$"+ 𝛽_%𝐼𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙𝑖𝑠𝑚 + 𝛽_&𝑃𝑜𝑤𝑒𝑟 𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 + 𝛽_'𝑀𝑎𝑠𝑐𝑢𝑙𝑖𝑛𝑖𝑡𝑦 + 𝛽₍𝑈𝑛𝑐𝑒𝑟𝑡𝑎𝑖𝑛𝑡𝑦 𝑎𝑣𝑜𝑖𝑑𝑎𝑛𝑐𝑒 + 𝛽₎𝐿𝑜𝑛𝑔 𝑡𝑒𝑟𝑚 𝑜𝑟𝑖𝑒𝑛𝑡𝑎𝑡𝑖𝑜𝑛 + 𝛽_*𝐼𝑛𝑑𝑢𝑙𝑔𝑒𝑛𝑐𝑒 + 𝛽₊𝐹𝑖𝑟𝑚 𝑠𝑖𝑧𝑒_#$"+ 𝛽_,𝑅𝑒𝑡𝑎𝑖𝑛𝑒𝑑 𝑒𝑎𝑟𝑛𝑖𝑛𝑔𝑠_#$"+ 𝛽_"!𝑃𝑟𝑜𝑓𝑖𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦_#$"+ 𝛽_""𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝑝𝑟𝑜𝑓𝑖𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦_#$"+ 𝛽_""𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝑙𝑒𝑣𝑒𝑟𝑎𝑔𝑒_#$" RESULTS

Leverage-Innovative culture model

Table 2 reports the result of Anderson-Hsiao estimators for model 1, random-effects regression for model 2 and model 3. Hypothesis 1 predicts that firms need adequate fund to continuously invest in knowledge-related assets in order to stay competitive in the market, which will place firms at a high leverage position. Model 1

suggests that the amount of intangible assets has a significant (p < 0.05) negative effect on leverage. The statistical significance has rejected hypothesis 1’s prediction, the result shows more possession of intangible assets resulted in lower leverage. Based on previous research, intangible assets are mostly classified as non-collateral in debt financing, therefore, firms that hold more intangible assets are facing a higher risk of bankruptcy when default. Although we argue that there are different forms of debt (e.g. convertible debt) can reduce the concerns, the result still consistent with the previous research (Long and Malitz, 1985).

Hypothesis 2 predicts a negative relationship between the cost of goods sold and leverage. It assumes that firms are motived to have a more efficient business process and lower costs of goods sold under an innovative environment. Firms require investments to optimize business process, which to some extent increases the firm’s need for debt financing. Model 1 exhibits a significant negative effect (p < 0.001) on leverage, which lower cost of goods sold reflects higher leverage. Therefore, hypothesis 2 is supported. Hypothesis 3 assumes that firms that are pursuing innovation on the basis of R&D will require ample fund, thus, firms will seek for higher leverage when retained earnings is not enough. After control for the firm’s retained earnings, model 1 shows that R&D intensity has a significant positive effect on leverage (P < 0.001), which hypothesis 3 is supported. Although both COGs and R&D intensity have a significant effect on leverage, the coefficients are not sizeable to explain the relationships. Therefore, we scale up the coefficients by divide both COGs and R&D intensity by 100,000. Besides intangible assets, COGs and R&D intensity have different

direction of effects on leverage, Model 1 suggests that innovative culture determines firms’ leverage from multiple dimensions.

The results for most control variables have observed significant effects. The positive effects of firms’ size and profitability are consistent with the previous study (Agca and Mozumdar, 2004). On the other hand, according to the pecking order theory,

firms prefer internal funding over debt financing. Firms will use less debt if they have more retained earnings, which the significant negative effect of retained earnings on leverage also seems logical (Myers and Majluf, 1984).

Individualism model

The potential positive impact of individualism on innovative culture is tested in model 2. Each dimension of the innovative culture model (Figure I) is tested separately. As predicted by hypothesis 4, there is a positive interaction between individualism and the amount of intangible assets (p < 0.001). The result suggests firms that located in higher individualism societies have higher intangible assets. The intangible assets essentially the hard-to-imitate resources that make firms differentiate from their rivals, a higher possession of intangible assets reflect a higher market orientation for innovation. Therefore, the result also implies a positive correlation between individualism and market orientation for innovation. In the aspect of strategic orientation for innovation, hypothesis 5 predicts that firms that located in more individualism societies are more advocate in a strategic investment in R&D. Model 2 exhibits a significant positive effect of individualism on R&D expenses (p < 0.01), which hypothesis 5 is supported. In contrast, hypothesis 6 is rejected by model 3 that individualism has no significant effects on COGs (p > 0.05), which implies that process innovation is indifferent between individualism societies and collectivism societies.

Model 3 is used to test for whether there is a direct effect of individualism on the firm’s leverage. Initially, we expect there is an indirect effect of individualism on leverage through the firm’s innovative culture based on the previous research. However, in model 2, we observe that individualism only has significant effects on two dimensions of the innovative culture (Figure I). It is not enough to conclude the relationship between individualism and innovative culture. Therefore, based on the result of model 1 and model 2, individualism does not have an indirect effect on firms’ leverage through innovative culture.

On the other hand, hypothesis 7 predicts that, in individualism societies, people are overconfidence and optimism in individualism societies, which could lead to overestimation of project profitability and debt capacity. The results from model 3 support hypothesis 7, individualism has a significant positive effect on leverage (p < 0.01). In other word, firms have higher leverage in more individualism societies. Therefore, instead of an indirect effect, individualism has a direct effect on leverage. Besides individualism, we also have included the other cultural dimensions of Hofstede’s index. The results of model 3 suggest that significant negative effect between long-term orientation and leverage, as well as a significant negative effect of uncertainty avoidance on leverage.

DISCUSSION AND CONCLUSION

the area of national culture, corporate culture and capital structure. Firstly, the previous research on the firm’s innovativeness is mostly focused on the strategic perspective that has not made a clear distinction between different dimensions. The strategic innovativeness, such as R&D investments cannot represent the overall innovativeness of a firm (Wang and Ahmed, 2004). We have proposed a model of innovative culture that includes market orientation for innovation, strategic orientation for innovation and process innovation. We have found a significant correlation between leverage and each of the dimension. Instead of a single effect direction on leverage, the multi-dimensional model has suggested both positive and negative effects on leverage. Although the relationship between innovative culture and the firm’s leverage cannot be simply stated, the opposite effect of each dimension on leverage actually contributes to determining the optimal leverage. Specifically, the results can be explained by the trade-off theory (Kraus and Litzenberger, 1973). Firms use debt to fund innovative investments not only produce hard-to-imitate resources that help them to stay competitive in competition, but firms also benefit from the deductibility of interest payment. However, the production of intangible assets places firms at higher cost and risk of bankruptcy as they are not tradeable when firms default (Long and Malitz, 1985). These factors help to determine how much debt firms are willing to use, which are consistent with the trade-off theory. The practical implication is that, firms tend to fund innovative investment with debt until the cost and risk of debt exceed the benefit.

Furthermore, there is a minor contribution of our study finds a positive relationship between individualism and firm’s leverage. The results are consistent with Antonczyk

and Salzmann (2014)’s finding, they have explained the positive relationship by the traits of overconfidence and optimism in individualism societies, which resulted in upward bias when evaluating the risk of debt and debt capacity. Model 3 also proposes that there are more potential effects of country culture on firm’s leverage, such as long-term orientation and uncertainty avoidance. Therefore, our paper provides some empirical evidence that country culture is an important determinant on firm’s leverage, which is also consistent with the preceding literature (Chui et al., 2016; Wang and Esqueda, 2014).

Alternative explanations

Although model 2 does not observe a significant effect of individualism and process innovation, there are relevant studies explain the relationship between individualism and process innovation in detail. According to Jang, Hong, Bock and Kim (2002), there is more process innovation in the Western countries as the culture is more market-oriented, the competition motivates people to seek for knowledge. In contrast, the East Asian culture is more relationship-centred and has fief-centred knowledge, which people prefer following well-documented paper. Market-oriented and relationship-centred culture are the traits in individualism societies and collectivism societies respectively (Hofstede, 1980).

If model 2 stands, we would then expect a potential indirect effect between individualism and leverage in future research, which firms have higher leverage in individualism societies because of having a more innovative culture.

Limitations

Our research is constrained by some limitations, which may affect the results. Firstly, in Compustat, there many firms do not report their R&D expenses frequently, which resulted in missing data for our primary variable. Although we assume these firms have no R&D activities by following the common practice in financial research, there are possibilities that we could overestimate or underestimate their efforts in R&D activities. Koh and Reeb (2015) have revealed that, in Compustat, 10% of firms with missing R&D data actually participant in innovation and R&D activities. As a consequence, missing data introduces bias to results. The missing data bias cannot be eliminated, however, results can still be improved by implementing an appropriate technique (Newman, 2014). Secondly, the debt market and financial policies are different across countries. For instance, some countries may have more debt options available and a lower barrier to debt financing (Beck, Demirguc-Kunt and Peria, 2006). Furthermore, the cost of bankruptcy also different across countries (Chui, Kwok and Zhou, 2016). These differences between countries may introduce some uncertainties to the relationship between individualism and firms’ leverage. Therefore, the future study should also control for the effects of GDP and the federal interest rate on the firm’s leverage at the macroeconomic level (Bokpin, 2009).

Although we construct the innovative culture from three different dimensions, admittedly, there are limitations of using observable financial outcomes to measure each dimension. The financial outcomes can reflect the innovativeness of a firm,

however, they may not be the best technique for comparison and generalization of an innovative culture. A survey index is recommended by Denison (1984), which measures personal perceptions on a firm’s innovativeness.

Future research

The future research can extend the innovative culture to more dimensions, the use of both survey index and financial outcomes may provide us with more reliable measurement on innovative culture. In addition, most of the results suggest the future study should extend to the effect of other country culture factors on leverage, as well as the relationship between country culture and corporate culture.

To conclude, the national culture has been coming more important in the age of global competition. It is important for managers to aware of the potential effect of national culture and corporate culture on the firm’s capital structure or even performance. The findings provide empirical evidence on the relationship between national culture, corporate culture and firm’s capital structure, it also extends innovativeness study on leverage to multi-dimensions.

Reference

Adrian, T., & Shin, H. S. (2010). Liquidity and Leverage. Journal of Financial Intermediation, 19(3), 418-437.

Agca, S., & Mozumdar, A. (2004). Firm Size, Debt Capacity, and Corporate Financing Choices. SSRN Electronic Journal. doi:10.2139/ssrn.687369

Antonczyk, R. C., & Salzmann, A. J. (2014). Overconfidence and optimism: The effect of national culture on capital structure. Research in International Business and Finance, 31, 132-151.

Ashenfelter, O., Levine, P. B., & Zimmerman, D. J. (2006). Statistics and Econometrics: Methods and Applications. John Wiley & Sons.

Bartoloni, E. (2011). Capital structure and innovation: Causality and determinants. Empirica, 40(1), 111-151.

Beck, T., Demirgüç-Kunt, A., & Soledad, M. P. (2006). Banking services for everyone? Barriers to bank access and use around the world. Washington, D.C.: World Bank.

Bokpin, G. A. (2009). Macroeconomic development and capital structure decisions of firms. Studies in Economics and Finance, 26(2), 129-142.

Campello, M. (2006). Debt financing: Does it boost or hurt firm performance in product markets? Journal of Financial Economics, 82(1), 135-172.

Chen, J. J. (2004). Determinants of capital structure of Chinese-listed companies. Journal of Business Research, 57(12), 1341-1351.

innovation. Pacific-Basin Finance Journal, 43, 173-187.

Chui, A. C., Kwok, C. C., & Zhou, G. (. (2016). National culture and the cost of debt. Journal of Banking & Finance, 69, 1-19.

Chui, A. C., Lloyd, A. E., & Kwok, C. C. (2002). The Determination of Capital Structure: Is National Culture a Missing Piece to the Puzzle? Journal of International Business Studies, 33(1), 99-127.

Czarnitzki, D., & Kraft, K. (2009). Capital control, debt financing and innovative activity. Journal of Economic Behavior & Organization, 71(2), 372-383. Denis, D. J., Denis, D. K., & Sarin, A. (1997). Agency Problems, Equity Ownership,

and Corporate Diversification. The Journal of Finance, 52(1), 135-160.

Denison, D. R. (1984). Bringing corporate culture to the bottom line. Organizational Dynamics, 13(2), 5-22.

Dierkens, N. (1991). Information Asymmetry and Equity Issues. The Journal of Financial and Quantitative Analysis, 26(2), 181.

Dobni, C. B. (2008). Measuring innovation culture in organizations. European Journal of Innovation Management, 11(4), 539-559.

Gill, A., Biger, N., & Mathur, N. (2011). The Effect of Capital Structure on Profitability: Evidence from the United States. International Journal of Management, 28(4). Gleason, K. C., Mathur, L. K., & Mathur, I. (2000). The Interrelationship between Culture, Capital Structure, and Performance: Evidence from European Retailers. Journal of Business Research.

Organizational Behavior and Human Decision Processes, 100(1), 96-109. Guiso, L., Sapienza, P., & Zingales, L. (2006). Does Culture Affect Economic

Outcomes? Journal of Economic Perspectives, 20(2), 23-48.

Hambrick, D. C., & Macmillan, I. C. (1985). Efficiency of Product R&D in Business Units: The Role of Strategic Context. Academy of Management Journal, 28(3), 527-547.

Harrison, L. E., & Huntington, S. P. (2000). Culture matters how values shape human progress.

Hofstede, G. (1980). Culture's consequences: International differences in work-related values. Newbury Park: Sage.

Hofstede, G. (2010). Dimensionalizing Cultures: The Hofstede Model in Context. Online Readings in Psychology and Culture, 2(1).

Hui, C. (1988). Measurement of individualism-collectivism. Journal of Research in Personality, 22(1), 17-36.

Hui, C. H., & Triandis, H. C. (1986). Individualism-Collectivism. Journal of Cross-Cultural Psychology, 17(2), 225-248.

Jang, S., Hong, K., Bock, G. W., & Kim, I. (2002). Knowledge management and process innovation: The knowledge transformation path in Samsung SDI. Journal of Knowledge Management, 6(5), 479-485.

Jansen, J. J., Van Den Bosch, F. A., & Volberda, H. W. (2006). Exploratory Innovation, Exploitative Innovation, and Performance: Effects of Organizational Antecedents and Environmental Moderators. Management Science, 52(11),

1661-1674.

Kayo, E. K., & Kimura, H. (2011). Hierarchical determinants of capital structure. Journal of Banking & Finance, 35(2), 358-371.

Koh, P., & Reeb, D. M. (2015). Missing R&D. Journal of Accounting and Economics, 60(1), 73-94.

Kraus, A., & Litzenberger, R. H. (1973). A State-Preference Model Of Optimal Financial Leverage. The Journal of Finance, 28(4), 911-922.

Lim, S. C., Macias, A. J., & Moeller, T. (2020). Intangible assets and capital structure. Journal of Banking & Finance, 118, 105873.

Long, M. S., & Malitz, I. B. (1985). Investment Patterns and Financial Leverage. Mitchell, D., & Coles, C. (2003). The ultimate competitive advantage of continuing

business model innovation. Journal of Business Strategy, 24(5), 15-21.

Modigliani, F., & Miller, M. H. (1958). The Cost of Capital, Corporation Finance and the Theory of Investment. The American Economic Review, 48(3), 261-297. Modigliani, F., & Miller, M. H. (1963). Corporate Income Taxes and the Cost of Capital:

A Correction. The American Economic Review, 53(3), 433-443.

Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics, 13(2), 187-221.

Naranjo‐Valencia, J. C., Jiménez‐Jiménez, D., & Sanz‐Valle, R. (2011). Innovation or imitation? The role of organizational culture. Management Decision, 49(1), 55-72.

Newman, D. A. (2014). Missing Data. Organizational Research Methods, 17(4), 372-411.

Nickell, S. (1981). Biases in Dynamic Models with Fixed Effects. Econometrica, 49(6), 1417.

Obrien, J. P. (2003). The capital structure implications of pursuing a strategy of innovation. Strategic Management Journal, 24(5), 415-431.

Olavarrieta, S., & Friedmann, R. (1999). Market-oriented culture, knowledge-related resources, reputational assets and superior performance: A conceptual framework. Journal of Strategic Marketing, 7(4), 215-228.

Oyserman, D., Coon, H. M., & Kemmelmeier, M. (2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128(1), 3-72.

Porter, M. E. (1986). Changing Patterns of International Competition. California Management Review, 28(2), 9-40.

Powell, D. M., Fu, R., Horowitz, K., Basore, P. A., Woodhouse, M., & Buonassisi, T. (2015). The capital intensity of photovoltaics manufacturing: Barrier to scale and opportunity for innovation. Energy & Environmental Science, 8(12), 3395-3408.

Rajan, R., Servaes, H., & Zingales, L. (2000). The Cost of Diversity: The Diversification Discount and Inefficient Investment. The Journal of Finance, 55(1), 35-80.

Organizational Dynamics, 12(2), 5-23.

Schwartz, H., & Davis, S. M. (1981). Matching corporate culture and business strategy. Organizational Dynamics, 10(1), 30-48.

Shane, S. (1993). Cultural influences on national rates of innovation. Journal of Business Venturing, 8(1), 59-73.

Shrivastava, P. (1985). Integrating Strategy Formulation with Organizational Culture. Journal of Business Strategy, 5(3), 103-111.

Simerly, R. L., & Li, M. (2000). Environmental dynamism, capital structure and performance: A theoretical integration and an empirical test. Strategic Management Journal, 21(1), 31-49.

Singh, J. (2008). Distributed R&D, cross-regional knowledge integration and quality of innovative output. Research Policy, 37(1), 77-96.

Thornton, P. H., Ocasio, W., & Lounsbury, M. (2015). The Institutional Logics Perspective.

Tsyplakov, S. (2008). Investment frictions and leverage dynamics☆. Journal of Financial Economics, 89(3), 423-443.

Verhees, F. J., & Meulenberg, M. T. (2004). Market orientation, innovativeness, product innovation, and performance in small firms. Journal of Small Business Management, 42(2), 134-154.

Wang, C. L., & Ahmed, P. K. (2004). The development and validation of the organisational innovativeness construct using confirmatory factor analysis. European Journal of Innovation Management, 7(4), 303-313.

Wang, D., & Esqueda, O. A. (2014). National cultural effects on leverage decisions: Evidence from emerging-market ADRs. Research in International Business and Finance, 31, 152-177.