Hành vi hối lộ, tiếp cận tín dụng, sức mạnh đàm phán của doanh nghiệp và sức ép cạnh tranh trên thị trường: Bằng chứng từ số liệu doanh nghiệp đa quốc gia

Chia sẻ: Vương Tâm Lăng | Ngày: | Loại File: PDF | Số trang:25

Thêm vào BST

Báo xấu

17
lượt xem 1
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

Nghiên cứu này sử dụng dữ liệu doanh nghiệp đa quốc gia, bao gồm 104 quốc gia trong giai đoạn từ 2010 tới 2019 để nghiên cứu tác động của hành vi đút lót tới việc tiếp cận tài chính của doanh nghiệp sở hữu sức mạnh đàm phán và/hoặc hoạt động trên thị trường có nhiều cạnh tranh. Mời các bạn cùng tham khảo!

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Hành vi hối lộ, tiếp cận tín dụng, sức mạnh đàm phán của doanh nghiệp và sức ép cạnh tranh trên thị trường: Bằng chứng từ số liệu doanh nghiệp đa quốc gia

INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 BRIBERY, CREDIT ACCESS, BARGAINING POWER AND MARKET COMPETITION: EVIDENCE FROM CROSS-COUNTRY FIRM-LEVEL DATA HÀNH VI HỐI LỘ, TIẾP CẬN TÍN DỤNG, SỨC MẠNH ĐÀM PHÁN CỦA DOANH NGHIỆP VÀ SỨC ÉP CẠNH TRANH TRÊN THỊ TRƯỜNG: BẰNG CHỨNG TỪ SỐ LIỆU DOANH NGHIỆP ĐA QUỐC GIA Dr. Le Thanh Ha; Dao Hanh Le; Nguyen Ngoc Mai National Economics University lethanhha@neu.edu.vn Abstract This paper used the multi-country firm-level data covering 104 countries for the period from 2010 to 2019 to investigate the effects of bribery on credit access for firms holding bargain- ing power and/or facing market competition. We used firms’ size and legal status to capture their bargaining power, while the levels of market competition were analyzed according to the number of competitors in the same working field. Our empirical results provided evidence to support the “greasing-the-wheels-of-credit access” hypothesis. Furthermore, the effects of bribery become stronger for larger-sized or formally-registered firms, and those facing no market competition. These effects also become pronounced if we controlled the endogeneity problem. Keywords: Bribery, Credit Access, Bargaining Power, Market Competition, Cross-country Firm-level. Tóm tắt Nghiên cứu này sử dụng dữ liệu doanh nghiệp đa quốc gia, bao gồm 104 quốc gia trong giai đoạn từ 2010 tới 2019 để nghiên cứu tác động của hành vi đút lót tới việc tiếp cận tài chính của doanh nghiệp sở hữu sức mạnh đàm phán và/hoặc hoạt động trên thị trường có nhiều cạnh tranh. Chúng tôi sử dụng kích thước doanh nghiệp và trạng thái pháp lý để ghi nhận sức mạnh đàm phán của doanh nghiệp, trong khi chúng tôi sử dụng thông tin về số lượng đối thủ cạnh tranh trên cùng lĩnh vực để đánh giá mức độ cạnh tranh trên thị trường mà doanh nghiệp hoạt động. Kết quả thực nghiệm chỉ ra bằng chứng ủng hộ giả thuyết bôi trơn để được tiếp cận các nguồn tài chính, tín dụng. Ngoài ra, tác động của hành vi đút lót trở nên mạnh hơn đối với doanh nghiệp có quy mô lớn hoặc doanh nghiệp đăng ký kinh doanh chính thức, hoặc doanh nghiệp không bị sức ép cạnh tranh quá lớn. Tác động cũng trở nên mạnh hơn khi cúng tôi kiểm soát vấn đề nội sinh. Từ khóa: Hối lộ, tiếp cận tín dụng, sức mạnh đàm phán của doanh nghiệp, cạnh tranh trên thị trường 1. 1. Introduction Undoubtedly, a capital plays an essential role for firms in contributing to the sustainable 7
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 growth and efficiency in firms’ operation activities, along with labors and technology (Aghion and Howitt, 2007). In addition to equity capital, firms use debt capital such as bank credit for their business activities. Accordingly, the access conditions to bank credit are important for the survival and development of both small-medium sized (SMEs) and large firms (Levine et al., 2000). Specifically, the micro and small firms make up for a large proportion of the developing economy (De Mel et al., 2009; Nichter and Goldmark, 2009) and financial constraints are con- sidered as a major obstacle for their daily business operations (Ayyagari et al., 2008; Beck and Demirguc-Kunt, 2008). However, the rigid financial system can create lots of challenges for firm to meets the conditions of financial institutions to issue the loans (Liu et al., 2020). Under this circumstance, firms may experience inefficient resource allocations and show the poor per- formance. In many countries, the government interferes in the economy by laws or regulations, there- fore, the quality of institutional environment not only plays an important role in credit market but also affect firms’ credit registrations (Galli et al., 2017). A great deal of papers studies the de- terminants of the quality of institutional environment such as performance in practicing laws (La Porta et al., 1997; Djankov et al., 2008). Firms may face difficulties to satisfy the requirements stemming from legal regulations requested by the government. Under this circumstance, paying bribes, with hardly any other choice is considered as a strategic behavior that allows firms to avoid the constraints from both the governments and banks (Zhou and Peng, 2012). Since bribery often happens in countries where the government involves in the economy (Klitgaard, 1988) or public servants possess low wages (Kraay and Van Rijckeghem, 1995), bribery can be served as a powerful tool for firms to help them to connect with public officials, then control their private benefits (Martin et al., 2007). In the literature, prior scholars have investigated the association between bribery and credit access. For example, Fungacova et al. (2015) provided empirical results that bribery has signif- icantly positive impacts on total firms’ bank debt in the short-term but negative impact in the long-term. Şeker and Yang (2014) claim that making informal payments such as bribery may cre- ate bigger burdens on the development of SMEs. Along with lending process, corruption can occur when firms bribe bank officials to achieving loans (Beck et al., 2006). In this spirit, bribing amount is seen as tax on firms, causing a barrier to get credit. In the same vein, Qi and Ongena (2018) and Wellalage and Locke (2019) indicates that bribery is detrimental for firms in accessing the credit. In contrast, some scholars have provided empirical results that bribery has a positive impact on firms’ chance of getting credit (Weill, 2011; Chen et al., 2013). It can be seen that find- ings of pre-existing papers have not consolidated about whether bribery has a positive or negative impact on firm’s credit access. Our paper also examines the relationship between bribery and firms’ credit access but crit- ically aims at closing the gap in the literature by two dimensions. First, we scrutinize to provide an apparent conclusion on the effects of bribery on firms’ credit access by using the cross-country firm level data as opposed to the empirical evidence in a typical country provided by previous studies. Second, we contribute to the literature by providing an argument that the impacts of bribery on credit access are conditional on firms’ bargaining power and level of market compe- tition. Put it differently, the effects of bribery are various among different situations. In terms of 8
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 bargaining power, in accordance with the analysis by Nguyen et al. (2020), we assume that a firm’s bargaining power is shaped by its competency and substitutes to bribes. Correspondingly, firms’ size and legality are applied to capture their bargaining power. Zhou and Peng (2012) de- bated that firms with weak bargaining power need to pay bribes for their survival, whilst firms with stronger bargaining power are more likely to achieve more benefits from their paid bribes (Rose-Ackerman, 1978). Based on this discussion, we formulate our hypothesis about effect of bribery on firm credit access under the moderation of firms’ bargaining power. In the aspect of level of market competition, it is believed that higher level of market competition leads to fewer benefits in firms’ paying bribes (Malesky et al., 2020). Under this circumstance, firms need to pay a higher amount of bribery in order to keep in touch with officials (Rose-Ackerman, 1978; Diably and Sylwester, 2015) and pay higher extra charges due to the less predictable environment (Malesky et al., 2020). In reference to this discussion, we develop our hypothesis about moder- ating effect of levels of market competition over the relationship between bribery and firms’ credit accessibility. In the present article, we employed a probit method for the firm-level cross-country data including 104 countries during the 2010-2019 period to examine the effects of bribery on firms’ credit access and our proposed hypotheses, which contemplates the moderating effects of firms’ bargaining power and the levels of market competition. We measured the bribery as the percentage of annual sales paid as informal payment to public officials, and generated dummy variable to twig the extent of bribery. In addition, we categorized firms by size as SMEs and large-sized firms based on the definition of World Bank Enterprise Survey. Dummy variable for firms’ legal status was also created. The levels of market competition were analyzed according to the number of competitors in the same working field. As mentioned in the literature, since there might be re- verse causality between bribery and credit accessibility, the bribery variable may be endogenous and lead to biased estimation results. Hence, we applied the instrumental variable (IV) method to address this endogeneity problem. To our best knowledge, our paper is the first to scrutinize the relationship between bribery and firms’ credit accessibility under the moderation of firms’ bargaining power and levels of mar- ket competition. The results of our benchmark model pointed out evidence to reinforce our hy- pothesis that firms’ bribery of public officials is positively related with firms’ credit access. Also, the estimation results of sub-sample by size and by legality also suggested that the effect of bribery is more sizable for larger-sized firms or formally-registered firms. The estimation results sub- sample by market competition demonstrate that the market competition negatively moderates the association between bribery and firms’ credit access. Lastly, the effect of bribery becomes stronger under the management for the endogeneity issue. The remainder of the paper is illustrated according to the following organizations. Related studies and hypothesis development are discussed in Section 2. Section 3 illustrates the data used in this paper. Section 4 performs the development of a model for enterprises at the cross-country firm-level data to determine a myriad of factors influencing firms’ credit accessibility. The em- pirical results are reported in Section 5. The conclusions and policy implications are provided in Section 6. 9
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 2. Theoretical background and hypothesis development 2.1. Bribery and firm credit access Bribery of public officials is more likely to happen in countries where there are govern- ment’s interventions in economy in terms of licenses or restrictions, which turns out to enhance the power of public officials in practicing regulations (Klitgaard, 1988). In addition, the rigid fi- nancial system can create lots of challenges for firm to meets the conditions of financial institu- tions to issue the loans (Liu et al., 2020). Under this circumstance, these firms need to select a strategic behavior to maintain the relationship with official in order to overcome these difficulties. As argued by Cai et al. (2011), bribery allow firms to obtain this goal. In this article, we expect that paying bribes increases firms’ accessibility to credit. There are plausible reasons to explain our belief. First, the government acts as a tangible hand in helping financial institutions to reduce problems about corporate moral hazard (La Porta et al., 2002; Faccio et al., 2006); thus, has strong impact on banks’ loans. As a result, firms can use distortive behaviors such as bribery to provoke public officials to abuse their reputations and authorities to control privately beneficial effects (Martin et al., 2007). For example, firms use bribery of public officials to take advantage over their competitors to obtain government contracts or be in governmental support programs as an award from the public officials for paying the highest bribery (Beck and Maher, 1986; Martin et al., 2007). This discussion implies that bribery can be served for buying judicial decisions or eliminating barriers that prevent firms from accessing to credit (Rose-Ackerman, 1998). Second, the red tape of financial institutions increases due to information asymmetry, which may lead to credit allocation inefficiency and higher transaction costs (Liu et al., 2020). Paying bribes can help firms to decrease the consequences of red tape and improve the likelihood of ob- taining bank loans (Fungacova et al., 2015). Additionally, Lui (1985) and Levine et al. (2000) contend that the rigidity of financial system can be softened, the cumbersome credit procedures can be simplified, the loan approval waiting time can be reduced for firms that accept to pay bribes. On the contrary, some scholars argue that bribery may be an obstacle in accessibility to credit of enterprises. It is illustrated that bank officers due to asymmetric information in lending process, bank officers often have permissions in determining credit terms like collateral types or interest rates (Barth et al., 2009), which leads to bribe from firms to break the barriers or avoid difficult terms and motivates officers to make up sophisticated credit terms to achieve more bribery (Guriev, 2004). This may lead to the fact that the number of non-performing loans in- creases due to firms’ bankrupt and reduce banks’ willingness to lend, which prevent firms from getting credit from banks. Additionally, banks’ severe credit terms to avoid default such as high interest rates make it harder for firm to get credit because of higher borrowing costs and more fi- nancial constraints (Beck et al., 2006; Firth et al., 2009), preventing low-risk borrowers, but en- couraging higher-risk borrowers to borrow and raising the risk of debt recovery. Hence, it is tougher for firms to access to credit. Overall, it is likely that bribery reduces credit constraints and helps firms obtain bank loans 10
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 whereas the relationship between banks and firms is promoted by corruption. Accordingly, we propose the following hypothesis: H1: Firms’ bribery of public officials is positively associated with firm credit access. 2.2. Moderating Effects of Bargaining Power Prior scholars such as Bliss and Tella (1997) and Fisman and Svensson (2007) contend that the bargaining power of firms determine the effects of bribery. In the similar spirit, we also argue that firms’ bargaining power influence their ability to reduce the payment amount and negotiate for better benefit (Rose-Ackerman, 1978), therefore moderate the association between bribery and credit access. In this paper, we employ firms’ size and legal status (formal and informal reg- istration) to capture their bargaining power vis-à-vis public officials. 2.3. Firm size and benefit of bribes As discussed by Rose-Ackerman (1978), the larger-sized firms are more likely to earn more benefit from their paid bribes. That is due to the fact that the local government prefers the large- sized firms since these firms with stronger financial and technical capacities tend to create more jobs and have more tax contributions. Furthermore, the large-sized firms have closed connections with political agents (Zhou and Peng, 2012), thus reduce the probability of paying “extortion” bribes and enhance the likelihood of accessing to more “lucrative business opportunities” (Nguyen et al., 2017). Hence, firms with a larger size are more likely to engage in benefit-seeking bribes and receive more preferential treatments. The more preferential treatments may help these firms to overcome the cumbersome credit procedures and reduce the loan approval waiting. Moreover, Wellalage et al. (2019) also reveal that the larger-size firms may be able to avoid the “loop-holes” and find an alternative way to address cumbersome situations. Moreover, transactional techniques based on verifiable hard information and more privilege lending relationships with larger and more transparent firms are more likely to be applied by large banks (Stein, 2002). Specifically, the firm size can impact the probability of bankruptcy since they are more diversified and harder to fail (Honhyan, 2009). That implies that they would have more opportunities to raise capital. Therefore, the firm size moderates the association between bribery and credit access. We hypoth- esize: H2: The effect of bribery on firm credit access is greater for large-sized firms. 2.4. Firm legality and benefit of bribes Informal (or unregistered) firms are seen potentially as productive as formal (or registered) firms. However, unregistered firms are believed to face constraints such as lack of access to formal loans or public services (Farazi, 2014), which prevents them from growing up (De Soto, 1989, 2000). Informal firms are likely to avoid taxes, however, have fewer chances to get access to formal loans or access to framework of legal protection supported by the government (Aureo de Paula and Scheinkman, 2007) or even government finance support due to institutional con- straints (Rothenberf et al., 2016). In this article, there are several reasons to make us believe that informal firms have very low bargaining power. First, since informal firms are usually recognized as very small firms (Farazi, 2014; Rothen- berf et al., 2016), they have lower bargaining power. Therefore, they earn fewer benefits from 11
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 their bribery actions as our argument about the association between firm size and benefits of bribes. Second, financial markets play important role in production activities both in the short- term and long-term investment. Their efficiency depends importantly on financial institutions, which connect lenders and borrowers by providing information to both sides at an acceptable re- payment rate. Hence, the financial institutions may reduce default and help to promote juridical credit return or guarantees (Pagano, 2001). This leads to the fact that these institutions require borrowers to provide sophisticated documents such as records about banking operations or fi- nancial statements, which tend to reject informal firms due to their lack of motivation to suffer additional costs for bank accounts or financial statements (Straub, 2005). Impossibility of bor- rowing formal loans raise the needs by informal firms either to find informal loans from informal resources or to pay bribes to bureaucrats to overcome the barriers created by financial institutions. They lead to higher costs for seeking credit and higher bribery costs due to uncertainty for infor- mal firms. Additionally, informal firms always participate in small trading, illegal market activities and tax nonpayment. Bases on the above discussion, we hold strong belief that legal registration positively moderate the relationship between bribery and firms’ credit access. We come up with the following hypothesis: H3: The effect of bribery on firm credit access is greater for formally-registered firms. 2.5. Moderating Effects of Market Competition In the literature, prior scholars have indicated two lines of thought guiding the firms’ bribery behavior, including the social norm view and the rent-seeking view. The former argues that the purpose of paying bribes regarded as an accepted norm in the business environment (Sundström, 2019) is to gain legitimacy and survive by conforming the accepted rules (DiMaggio and Powell, 1983), while the latter emphasizes on the expected abnormal rents (Rose-Ackerman, 1978). Al- though there are differences in explaining the firm’s bribery behavior, these two streams have af- firmed that the increased competition among firms affects the choice of whether firms engage in bribery. In this article, we argue that the rise in firms’ competition in the market can influence the effects of bribery on firms’ ability in accessing credit. The reasons are as follows. First, according to the social norm view, the increase in competitors in the operating environment where bribery is accepted as a normal social norm may cause firms to pay more bribes to maintain the relation- ship with the bureaucrats (Malesky et al., 2020). These higher bribe amounts stem from a reduc- tion in the firm bargaining power vis-à-vis to public officials due to the highly competitive business environment (Rose-Ackerman, 1978). Moreover, Galang (2012) states that a diminish in firms’ bargaining power also implies that firms may receive less preferential treatments and greater transaction costs. Therefore, the advantages over competitors because of paying bribes can reduce in the more competitive business environment. Second, we follow North (1990) and Williamson (2000) to base on the rent-seeking view to contend that the impacts of market com- petition on the relationship between bribery and firms’ credit access are conditional on the pre- dictability level of policy. Specifically, firms have more opportunities to access credit if the environment is less predictable (Galang, 2012; Zhou et al., 2013). Conversely, firms in the less 12
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 predictable environment may have to pay additional costs (Malesky et al., 2020). The market competition leads to more policy uncertainty (Alexeev and Song, 2013), thus influences the pre- dictability level of policy implementation. Accordingly, the firms’ accessibility of credit is more likely to decrease in the highly competitive business environment. Therefore, we hypothesize: H4: The market competition negatively moderates the association between bribery and firms’ credit access. We summarize our theoretical framework in Figure 1. Figure 1: Theoretical framework. 3. Data Description This paper used cross-sectional data from the World Bank Enterprise Survey. The data cleaning was carried out by getting rid of missing observations and outliers and keeping the data of manufacturing firms corresponding to trade theory. There are total 24,661 observations cov- ering 104 countries for the period from 2010 to 2019. Dependent variable: CA CA is a dichotomous variable, which takes the value of 1 if a firm get credit access and 0 otherwise. There are about 38.86% of firms getting credit in our sample. Key independent variable: DBri and Bri. Bri is the natural logarithm of the percentage of annual sales paid as informal payment to public officials, and DBri is the dichotomous variable that takes the value of 1 if a firm involves in bribery activities, and 0 otherwise. Table 1 shows that there are hardly any differences between non-credit-access firms and credit-access firms in using bribery in our database. 13
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 Table 1: Comparisons of credit access vs. non-credit access Variable Non-CA CA DBri 0.115 0.117 Bri 0.193 0.186 Briamount 1.262 2.512 ource: Authors’ calculation In Table 2, the data were separated into various groups to analyze the moderating effects of bargaining power on the relationship between bribery and credit access. Institutional constraints used in this paper is gathered from firms’ opinion about the level of obstacles influencing their business: corruption, political instability, tax administrations, and business licensing and permits. Panel A indicates that larger firms with higher bargaining power are more likely to get ac- cess to credit. For “Credit Access only” category, unlike 23.94% of small firms and 35.19% of medium firms, 46.25% of large-sized firms access to credit. Also, for “credit access and bribery”, larger firms still have higher percentage. However, for “bribery only”, large-sized firms account for only 4.99% while small and medium firms account for 8.93% and 7.43% respectively, which implies that larger firms use less bribe. In the following analysis, we concentrate on firms’ legal status. As described in Table 2, the proportion of informal firms getting access to credit is lower, but the fraction of paying-bribe firms is higher than the formal ones. Roughly 23% of informal firms get access to credit, but about 33.96% of formal firms get access to credit without bribery. Also, for “bribery without credit access”, although informal firms account for 11.86%, the formal ones only account for 7.32%, which implies that informal firms need to bribe not only for credit access but also for other issues. Panel B describes firms with and without market competition and level of market compe- tition associated with bribery and credit access. It can be observed that firms without market com- petition have more opportunities in accessing to credit than those with market competition (47.82% comparing to 38.81%). However, bribery hardly changes along with market competition. For level of market competition, it is also shown that firms with lower competition-level are easier in credit access (60.37% comparing to 33.69%) and bribery remains stable in this situation. Table 2: Percentages of credit access and bribery firms (%) Panel A: Bargaining power (%) CA & Bribery Small Medium Large Total Formal Informal Total CA only 23.94 35.19 46.25 33.86 33.96 23.31 33.86 CA and bribery 4.02 5.28 6.09 5.02 5.04 3.39 5.02 Bribery only 8.93 7.43 4.99 7.37 7.32 11.86 7.37 No CA & no bribery 63.11 52.10 42.66 53.75 53.68 61.44 53.75 Total 100 100 100 100 100 100 100 14
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 Panel B: Market competition (%) Market Competition CA & Bribery No Yes Total Low High Total CA only 43.15 33.78 33.86 52.70 29.30 33.86 CA and Bribery 4.67 5.03 5.02 7.67 4.39 5.02 Bribery only 7.61 7.37 7.37 5.02 7.94 7.37 No CA & no bribery 44.67 53.83 53.75 34.62 58.38 53.75 Total 100 100 100 100 100 100 Source: Authors’ calculation Control variables Our control variables used in this paper are relied on the list of control variables made by Wellalage et al. (2020). Annual sale per labor () refers to firm’s business performance during a fiscal year. As firms with better performance are likely to pay back loans easier, they have higher chance to get credit. and are the firm’s age and the experience in this firm’s working sector. Older firms are more knowledgeable; hence, they are easier in approaching credit. We also ex- pected that higher managing experience helps firms to deal with credit accessing processes and raise the chance to obtain credit. Size, expressing the scale-effect, displays the number of em- ployees. Larger firms are predicted to have higher ability in credit access. and represents the percent of a firm owned by the Government (or State) and by foreign companies, respectively. takes the value of 1 if a firm involves in direct export activities and 0 otherwise. takes the value of 1 if a firm carried out research and development activities and 0 otherwise. The summary sta- tistics for all sample and sub-sample are described in Table 3. Table 3: Control variables All sample CA=0 CA=1 Obs Mean Std. Dev Obs Mean Std. Dev Obs Mean Std. Dev LnSalecap 24,661 13.43 2.63 15073 13.14 2.47 9588 13.90 2.81 LnAge 24,661 2.80 0.77 15073 2.72 0.77 9588 2.92 0.77 LnManager 24,661 2.82 0.65 15073 2.75 0.66 9588 2.92 0.63 Size 24,661 112.28 237.66 15073 88.22 206.48 9588 150.116 375.528 Export 24,661 0.23 0.42 15073 0.17 0.37 9588 0.34 0.47 RD 24,661 0.27 0.44 15073 0.20 0.40 9588 0.38 0.48 State 24,661 0.00 0.02 15073 0.00 0.02 9588 0.00 0.01 Foreign 24,661 0.06 0.22 15073 0.06 0.22 9588 0.07 0.23 Source: Authors’ calculation 15
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 The correlation matric between the key variables in this paper is described in Table 4. Most of the correlation coefficients are seen to be lower than 0.4. Moreover, the higher-than-0.4 correlation coefficient belongs to those between dichotomous variables about firms’ bribe paying and the bribe payment amount. Hence, there is no problem with multicollinearity in our regressions. Table 4: Correlation matrix DBri Bri Export RD LnSalecap State Foreign LnAge LnManager Size DBri 1.00 Bri 0.88 1.00 Export 0.00 0.00 1.00 RD 0.00 0.01 0.16 1.00 LnSalecap -0.02 -0.03 0.02 0.08 1.00 State 0.02 0.03 0.03 0.00 -0.00 1.00 Foreign 0.01 0.01 0.21 0.06 0.03 0.01 1.00 LnAge -0.01 -0.02 0.13 0.10 0.07 0.03 0.00 1.00 LnManager -0.02 -0.03 0.09 0.04 -0.00 -0.02 -0.00 -0.39 1.00 Size -0.01 - 0.01 0.31 0.16 0.04 0.09 0.21 0.17 0.04 1.00 Source: Authors’ calculation 4. Model Specification The benchmark model is specified as follows: CAi = βo + β1Briberyi + β2CONTROLi + voc + λt + εi (1) where subscript i, c, and t denote firm, country, and year, respectively. and are, in turn, country and year fixed effects. is the credit access of firm i. is a set of bribery variables. is a set of control variables. is an error term. As is a binary variable, we applied the probit method to es- timate equation (1). We reported the marginal effects at the mean level. The country fixed effects, , were used to reflect the unobservable factors that are specific to countries. In addition, as the data in our sample were collected in different years, the year fixed effects were used to capture the macroeconomic variables of the world economy which change over time and affect all firms. We first tested the association between bribery and firm credit access. Since we proposed that this relationship may depend on a firm’s bargaining power, we re-regressed the equation (1) in the sub-sample based on firm size and legal status. In addition to a firm’s bargaining power, we also investigated the moderating effects of market competition. We compared the effects of bribery on credit access probability for the sub-sample of firms facing competition of others in the market and those having no market competition. For further analysis, we also consider the 16
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 effects of market competition level. In particular, we define the low and high level of market competition and then also re-estimated the equation (1) with the sub-sample of each level. The interactions between market competition and the firm’s size were also examined. In the further analysis, we have investigated the effect of bribery on firm credit access across sectors. In our model, there is a simultaneity between bribery and firm credit access that may cause our results to be biased. On the one hand, paying bribes increases firms’ accessibility to credit. According to Rose-Ackerman (1998), firms use bribery to buy judicial decisions or eliminate barriers preventing firms from credit access. Banks or financial institutions may expand the ability of credit access for firms who are willing to pay bribes. Moreover, paying bribes can help firms to decrease the consequences of information asymmetry and improve the likelihood of obtaining bank loans (Fungacova et al., 2015). On the other hand, corruption encourages credit constraints and built the impediments of firm financing (Qi and Ongena, 2019). This is on the grounds that in a corrupt lending process, bank officials pick firms who are ready to pay bribes instead of credible ones. Difficulties get worse for SMEs that are contingent upon limited financial re- sources. These discussions support the view that there is an association between firm credit access and bribery. To mitigate the consequences, we used the instrumental variable (IV) method to address the endogeneity bias between the credit constraint proxy and DBri and Bri variables. Following Fisman and Svensson (2007) and Nguyen et al. (2017), the sector-location average approach is employed to fix the issue of endogeneity bias. According to Qi and Ongena (2018), firm bribery is averaged across other firms within the same locality and sector but excludes the firm itself. As a result, the sector-location average of bribery payment amount, which is exogenous to the firm, specified by the business method of the sector and the rent-seeking ability of the bureaucrats. More clearly, instrumenting bribery by the bribery payment amount denoted by GBri_ivb and GBri_ivr can minimize omitted unobservable errors related with bribe intensity at the firm, but not the sector level. Johnson et al. (2000) utilize the judiciary system as a proxy for the quality of property rights since fairness and how it could affect business is another instrument for bribery. Accordingly, this study uses Weak_judiciary as an instrument for Bri due to firm credit access depending on the degree of judiciary system fairness. The instrument Weak_judiciary refers to firm’s attitude toward level of judiciary system corruption, based on a question: “The court system is fair, impartial and uncorrupted”. 5. Estimation Results 5.1. Main Results Column 1-6 of Table 5 presents the estimation results of bribery on firm credit access. More specifically, loans of firms from distinct sources, including private banks and from state banks are shown in columns 1-2, 3-4 and 5-6, respectively. In this analysis, we use the dummy variable that takes the value of 1 if firms pay bribes (DBri) and the bribery payment amount that is a share of revenue (Bri) paid as informal payment to public officials. The results show that there is a pos- itive relationship between bribery and firm credit access. Firms’ bribery of public officials in- creases the probability of credit access to both private and public banks by 0.23%. Engaging in greasing bribery to get things done with regard to licenses, taxes, regulations, or services in private 17
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 banks increases the probability of accessing to external finance from private banks by 0.20%. We also observe the similar evidence when using the bribery payment amount. In particular, the higher share of sales paid as greasing bribery increases the probability of credit access by 9% and 8% in all banks and in private banks, respectively. These findings are consistent with those of Chen at al. (2013) and Liu et al. (2020), who observed a positive link between the amout of bribes and external credit access. Li et al. (2008) and Faccio (2010) also provide the similar find- ing and contend that firms can use bribery to overcome market failures and eliminate ideological discriminations to access banks loans. However, the result of bribery on firm’s accessing to fi- nance provided by state banks is statistically insignificant in our database. Regarding the effects of other control variables on firm credit access, a statistically signif- icant positive relationship is found between exporting firms and credit access. The table reports that export firms are approximately 0.32% less credit constraint than non-exporters. The result is the same for firms having R&D activities. The sale intensity per labor (Lnsalecap) and the firm age (Lnage) are positive and statistically significant, whereas the foreign linkages (Foreign) and the relationship with state sectors (State) reduce the probability of firms’ credit access in our the- oretical model. That is due to the fact that firms having relationships with state sectors often have more external financing channels than those without this connection since they can achieve more government shelters and soft budget constraints. Furthermore, the results also indicate a positive link between the firm size measured by the number of full-time employees (Size) and their credit accessibility, which aligns with Rose-Ackerman (1978). As revealed by Gonzalez (2015), small firms often encounter more financing constraints and have difficulties in accessing the formal fi- nancial system, therefore banks’ preferences have larger effects on them. Our results are aligned with Liu et al. (2020). The positive coefficient of LnAge implies that older firms are more likely to get access to finance as a result of knowledge and experience accumulation over the years. The more experienced managers have higher capacities of negotiating and persuading partners, which explains why the relationship between the working experience of top managers and credit access in all banks except for private banks is positive. Table 5: Benchmark estimation results (1) (2) (3) (4) (5) (6) VARIABLES All Private Banks State Banks DBri 0.23*** 0.20*** 0.07 Bri 0.09*** 0.08*** 0.01 (0.017) (0.018) (0.028) (0.031) (0.034) (0.047) RD 0.32*** 0.32*** 0.27*** 0.27*** 0.22*** 0.22*** (0.022) (0.022) (0.025) (0.025) (0.030) (0.030) Export 0.32*** 0.32*** 0.26*** 0.26*** 0.27*** 0.27*** (0.024) (0.024) (0.026) (0.026) (0.034) (0.034) 18
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 LnSalecap 0.10*** 0.10*** 0.13*** 0.13*** 0.04*** 0.04*** (0.007) (0.007) (0.008) (0.008) (0.010) (0.010) State -0.44 -0.42 -1.19** -1.16** 0.87 0.88 (0.449) (0.447) (0.536) (0.535) (0.566) (0.566) Foreign -0.40*** -0.40*** -0.33*** -0.34*** -0.46*** -0.46*** (0.045) (0.044) (0.047) (0.047) (0.079) (0.079) LnAge 0.04*** 0.04*** 0.06*** 0.06*** -0.02 -0.02 (0.014) (0.014) (0.015) (0.015) (0.020) (0.020) LnManager 0.07*** 0.07*** -0.01 -0.01 0.19*** 0.19*** (0.016) (0.016) (0.018) (0.018) (0.025) (0.024) Size 0.00*** 0.00*** 0.00*** 0.00*** 0.00 0.00 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) Constant -2.21*** -2.20*** -2.57*** -2.56*** -3.11*** -3.10*** (0.186) (0.186) (0.209) (0.209) (0.340) (0.339) Observations 23,317 23,317 23,281 23,281 22,062 22,062 Pseudo R2 0.196 0.195 0.307 0.307 0.186 0.186 Robust standard errors in parentheses *** p
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 Table 6: Estimation results of sub-sample by size VARIABLES (1) (2) (3) (3) SMEs Large SMEs Large DBri 0.18*** 0.38*** (0.035) (0.069) Bri 0.09*** 0.14*** (0.019) (0.040) RD 0.27*** 0.33*** 0.27*** 0.33*** (0.027) (0.040) (0.027) (0.040) Export 0.18*** 0.31*** 0.18*** 0.32*** (0.032) (0.040) (0.032) (0.039) LnSalecap 0.10*** 0.08*** 0.10*** 0.08*** (0.009) (0.014) (0.009) (0.014) State -0.80 -0.97* -0.82 -0.90 (0.772) (0.581) (0.772) (0.580) Foreign -0.43*** -0.54*** -0.43*** -0.54*** (0.068) (0.063) (0.068) (0.063) LnAge -0.03** 0.11*** -0.03** 0.11*** (0.017) (0.027) (0.017) (0.027) LnManager 0.09*** 0.09*** 0.09*** 0.09*** (0.019) (0.032) (0.019) (0.032) Size 0.01*** 0.00** 0.01*** 0.00** (0.000) (0.000) (0.000) (0.000) Constant -2.33*** -1.65*** -2.33*** -1.63*** (0.231) (0.359) (0.231) (0.358) Observations 17,446 5,860 17,446 5,860 Robust standard errors in parentheses *** p
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 formally registered. Regarding other control variables, they are only statistically significant in the sub-sample of registered firms. These empirical results support our proposed hypothesis H3 that the effect of bribery on firm credit access is greater for formally-registered firms. Table 7: Estimation results of sub-sample by legality (1) (2) (3) (4) No Yes No Yes VARIABLES CA CA CA CA DBri 0.07 0.23*** (0.405) (0.031) Bri -0.02 0.09*** (0.196) (0.017) RD -0.35 0.32*** -0.33 0.32*** (0.400) (0.022) (0.398) (0.022) Export 0.14 0.33*** 0.12 0.33*** (0.410) (0.024) (0.410) (0.024) LnSalecap -0.05 0.10*** -0.05 0.10*** (0.087) (0.007) (0.086) (0.007) State -6.66 -0.44 -6.70 -0.42 (6.443) (0.449) (7.148) (0.448) Foreign 0.47 -0.40*** 0.46 -0.40*** (0.988) (0.045) (0.984) (0.045) LnAge 0.13 0.04*** 0.14 0.04*** (0.147) (0.014) (0.146) (0.014) LnManager 0.03 0.07*** 0.04 0.07*** (0.163) (0.016) (0.163) (0.016) Size 0.00** 0.00*** 0.00** 0.00*** (0.001) (0.000) (0.001) (0.000) Constant 0.35 -2.20*** 0.35 -2.19*** (1.659) (0.187) (1.644) (0.187) Observations 168 23,089 168 23,089 Pseudo R2 0.312 0.195 0.312 0.194 Robust standard errors in parentheses *** p
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 In the next step, we investigate the important role of market competition that may influence the association between bribery and firm credit access. The effects of greasing bribery on firm’s ability in attracting venture capital funding probability are stimulated if there is no competition within the markets. The impacts of greasing bribes increase from 0.24% to 0.97% for firms that do and do not face competitors in selling their main products, respectively. The similar evidence can be found when considering the effects paying bribery payment amounts to bureaucrats. Fur- thermore, Table 8 also highlights the differences between firm credit access probability and the level of market competition. We define the market as the low competition if firms have less than 5 competitors and as the high competition if firms have more than 5 competitors1. The higher the number of competitors in the market, the harder it is for firms to access external finance when paying bribes. The role of bribery in promoting the firms’ ability in getting external finance be- comes less sizable if they operate in the highly competitive markets. The results in Table 8 rein- force the conclusion that the market competition negatively moderates the association between bribery and firms’ credit access. Table 8: Bribery and credit access with market competition (2) (2) (3) (4) (5) (6) (7) (8) No Yes No Yes Low High Low High VARIABLES CA CA CA CA CA CA CA CA DBri 0.97** 0.24*** 0.32*** 0.22*** (0.390) (0.032) (0.068) (0.036) Bri 0.48*** 0.09*** 0.12*** 0.09*** (0.178) (0.017) (0.034) (0.020) RD 0.31 0.32*** 0.32 0.32*** 0.26*** 0.34*** 0.26*** 0.34*** (0.266) (0.022) (0.266) (0.022) (0.043) (0.026) (0.043) (0.026) Export 0.41 0.32*** 0.40 0.33*** 0.29*** 0.33*** 0.29*** 0.33*** (0.324) (0.025) (0.326) (0.025) (0.054) (0.028) (0.054) (0.028) LnSalecap 0.29** 0.10*** 0.31*** 0.10*** 0.14*** 0.10*** 0.14*** 0.10*** (0.117) (0.007) (0.118) (0.007) (0.022) (0.008) (0.022) (0.008) Foreign -1.95*** -0.39*** -1.98*** -0.39*** -0.48*** -0.37*** -0.48*** -0.38*** (0.571) (0.046) (0.575) (0.046) (0.091) (0.054) (0.091) (0.054) 1. Based on the questionnaire about the number of competitors, we select 5 competitors to define the low and high level of competition. 22
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 (0.167) (0.014) (0.167) (0.014) (0.029) (0.016) (0.029) (0.016) LnManager -0.03 0.07*** -0.04 0.07*** -0.01 0.10*** -0.01 0.10*** (0.173) (0.017) (0.174) (0.017) (0.035) (0.019) (0.035) (0.019) Size 0.00 0.00*** 0.00 0.00*** 0.00*** 0.00*** 0.00*** 0.00*** (0.001) (0.000) (0.001) (0.000) (0.000) (0.000) (0.000) (0.000) State -0.70 -0.68 -3.00 -0.51 -3.02 -0.49 (0.474) (0.473) (2.053) (0.478) (2.045) (0.477) Constant -2.72* -2.21*** -2.90* -2.20*** -2.08*** -2.27*** -2.06*** -2.26*** (1.646) (0.189) (1.653) (0.189) (0.372) (0.192) (0.371) (0.192) Observations 184 22,297 184 22,297 4,759 17,728 4,759 17,728 Pseudo R2 0.378 0.201 0.383 0.200 0.175 0.171 0.174 0.170 Robust standard errors in parentheses *** p
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 Table 9: Bribery and credit access with interaction between firms’ size and firms facing market competition (1) (2) (3) (4) (5) (6) (7) (8) Low Market Competition High Market Competition Low Market Competition High Market Competition SMEs Large SMEs Large SMEs Large SMEs Large VARIABLES CA CA CA CA CA CA CA CA DBri 0.29*** 0.77*** 0.15*** 0.35*** (0.072) (0.267) (0.041) (0.072) Bri 0.11*** 0.42** 0.08*** 0.12*** (0.034) (0.177) (0.023) (0.041) RD 0.23*** 0.17* 0.29*** 0.36*** 0.23*** 0.17* 0.29*** 0.37*** (0.049) (0.098) (0.032) (0.044) (0.049) (0.098) (0.032) (0.044) Export 0.16** 0.23** 0.18*** 0.33*** 0.16** 0.23** 0.18*** 0.33*** (0.067) (0.098) (0.038) (0.044) (0.067) (0.098) (0.038) (0.044) LnSalecap 0.12*** 0.10* 0.09*** 0.08*** 0.12*** 0.10* 0.09*** 0.08*** (0.025) (0.053) (0.010) (0.014) (0.025) (0.052) (0.010) (0.014) State -3.71 -4.19* -0.62 -0.84 -3.69 -4.23* -0.64 -0.77 (3.713) (2.471) (0.767) (0.590) (3.692) (2.468) (0.767) (0.588) Foreign -0.61*** -0.49*** -0.37*** -0.58*** -0.62*** -0.50*** -0.37*** -0.59*** (0.130) (0.132) (0.080) (0.074) (0.130) (0.131) (0.080) (0.074) LnAge -0.08** -0.01 -0.02 0.13*** -0.08** -0.01 -0.02 0.13*** (0.033) (0.066) (0.019) (0.030) (0.033) (0.066) (0.019) (0.030) LnManager -0.02 0.10 0.13*** 0.10*** -0.02 0.09 0.13*** 0.11*** (0.040) (0.077) (0.022) (0.036) (0.040) (0.076) (0.022) (0.036) Size 0.01*** 0.00*** 0.01*** 0.00 0.01*** 0.00*** 0.01*** 0.00 (0.001) (0.000) (0.001) (0.000) (0.001) (0.000) (0.001) (0.000) Constant -2.08*** -1.21 -2.37*** -1.71*** -2.07*** -1.17 -2.37*** -1.70*** (0.431) (0.876) (0.235) (0.365) (0.431) (0.871) (0.235) (0.365) Observations 3,688 1,112 13,724 4,734 3,688 1,112 13,724 4,734 Pseudo R2 0.162 0.230 0.161 0.174 0.161 0.228 0.161 0.172 Robust standard errors in parentheses *** p
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 5.2. Sector Variation In this part, we examined the effect of bribery on firm credit access in different sectors. We divided the data into three sectors as in Tomiura (2007): supplier-dominated, scale-intensive, and science-based. There are 9,855 firms in supplier-dominated sector, 4,723 firms in scale-in- tensive sector, and 7,184 firms in science-based sector. The effect of bribery on credit access in science-based sector is the largest, while the ones in supplier-dominated and scale-intensive sector remain quite similar to each other. Table 10 also shows that there is no significant difference in effects of bribery on the probability of accessing credit across sectors. Table 10: Estimation results of sub-sample by sector Supplier-dominated Scale-intensive Science-based VARIABLES CA CA CA CA CA CA DBri 0.21*** 0.21*** 0.24*** (0.050) (0.072) (0.058) Bri 0.09*** 0.08* 0.09*** (0.026) (0.039) (0.032) RD 0.32*** 0.32*** 0.29*** 0.29*** 0.37*** 0.37*** (0.035) (0.035) (0.052) (0.052) (0.038) (0.038) Export 0.32*** 0.32*** 0.37*** 0.37*** 0.24*** 0.24*** (0.037) (0.037) (0.060) (0.060) (0.044) (0.044) LnSalecap 0.10*** 0.10*** 0.08*** 0.08*** 0.12*** 0.12*** (0.011) (0.011) (0.017) (0.017) (0.014) (0.014) State 0.13 0.14 -1.64 -1.66 -0.13 -0.08 (0.651) (0.648) (1.236) (1.234) (0.828) (0.827) Foreign -0.40*** -0.40*** -0.34*** -0.34*** -0.46*** -0.46*** (0.069) (0.068) (0.109) (0.109) (0.080) (0.080) LnAge 0.02 0.02 0.00 0.00 0.07*** 0.08*** (0.021) (0.021) (0.034) (0.034) (0.026) (0.026) LnManager 0.05** 0.05** 0.12*** 0.12*** 0.07** 0.06** (0.025) (0.025) (0.039) (0.039) (0.031) (0.031) Size 0.00*** 0.00*** 0.00*** 0.00*** 0.00*** 0.00*** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) Constant -2.06*** -2.06*** -1.23** -1.21** -2.79*** -2.78*** (0.268) (0.268) (0.547) (0.547) (0.422) (0.420) Observations 9,421 9,421 4,373 4,373 6,652 6,652 Pseudo R2 0.197 0.197 0.186 0.185 0.200 0.199 Robust standard errors in parentheses *** p
INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 5.3. IV Estimation Until now, we have not dealt with the endogeneity issue that might cause our estimation results to be biased. We, therefore, concentrate on this issue in this part. We follow Wellalage et al. (2020) to use the sector-country average bribing amount (GBri_ivb and GBri_ivr) and firm’s attitude toward the level of judiciary system corruption (Weak_Judiciary) as our instruments. We then re-investigate the effects of bribery on the probability of credit access by employing instru- mental variable probit models. We construct two sets of instruments: (GBri_ivb and Weak_Judi- ciary) for DBri and (GBri_ivr and Weak_Judiciary) for Bri. The results of the IV regression on full sample and sub-sample are displayed in Table 11. Comparing to the benchmark estimation results in Table 5, our hypothesis about the effects of bribery in credit access remains unchanged. All coefficients’ signs stay steadily with our earlier estimations, but the marginal effects are substantially greater. These results firmly reinforce our argument about the existence of “greasing the wheel of credit access” hypothesis. Overall, the results coming from this IV model estimation strengthen our main conclusion that bribery has positive effect on firm credit access. When contemplating endogeneity issue, the influence of bribery on credit access is notably greater. Table 11: IV model estimation (1) (2) (5) (6) VARIABLES CA GDBri CA GBri GDBri 0.45*** (0.119) GBri 0.17*** (0.062) RD 0.32*** 0.01*** 0.32*** 0.03*** (0.023) (0.005) (0.023) (0.009) Export 0.33*** 0.02*** 0.33*** 0.03*** (0.025) (0.005) (0.025) (0.010) LnSalecap 0.11*** -0.01*** 0.11*** -0.02*** (0.007) (0.002) (0.007) (0.003) State -0.61 0.10 -0.58 0.17 (0.472) (0.096) (0.472) (0.180) Foreign -0.36*** -0.02** -0.37*** -0.02 (0.045) (0.009) (0.045) (0.017) LnAge 0.03** 0.00 0.04** 0.00 (0.014) (0.003) (0.014) (0.005) LnManager 0.08*** -0.01*** 0.08*** -0.02*** (0.017) (0.003) (0.017) (0.006) 26