Bribery-export nexus under the firm's growth obstacles

Trang Hoai Phan; Stachuletz, Rainer Article Bribery-export nexus under the firm's growth obstacles Economies Provided in Cooperation with: MDPI – Multidisciplinary Digital Publishing Institute, Basel Suggested Citation: Trang Hoai Phan; Stachuletz, Rainer (2022) : Bribery-export nexus under the firm's growth obstacles, Economies, ISSN 2227-7099, MDPI, Basel, Vol. 10, Iss. 2, pp. 1-26, https://doi.org/10.3390/economies10020028 This Version is available at: https://hdl.handle.net/10419/257391 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/   Citation: Phan, Trang Hoai, and Rainer Stachuletz. 2022. Bribery—Export Nexus under the Firm’s Growth Obstacles. Economies 10: 28. https://dx.doi.org/ 10.3390/economies10020028 Academic Editor: George R.G. Clarke Received: 19 November 2021 Accepted: 12 January 2022 Published: 19 January 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). economies Article Bribery—Export Nexus under the Firm’s Growth Obstacles Trang Hoai Phan 1,* , Rainer Stachuletz 2 1International Economics, Technical University of Darmstadt, Hochschulstr. 1, 64289 Darmstadt, Germany 2 Department of Economics 1, Berlin School of Economics and Law, Badensche Str. 52, 10825 Berlin, Germany; rainer[email protected] *Correspondence: [email protected]; Tel.: +49-6151-16-22872 Abstract: Business bribery is a particularly serious problem in the integration era. First, this article investigates the effects of institutional obstacles on firms’ bribery in 131 countries classified by nation income groups. Through the appropriate proposal of fitting functions, the relationship between obstructions and the predicted margin effect of bribery is intuitively elucidated. Second, this paper sheds light on the relationship between bribery payment and exports. Then the analysis is upgraded when controlling for the moderation of a firm’s growth constraints. The results detected that not only institutional barriers, but also internal and external hindrances play an essential role in the interaction between bribe payments and export share. More interestingly, this study scrutinizes the role of obstacles in this relationship separately. Besides, SMEs and large enterprises are also adopted in further sensitivity analyses. To solve the endogeneity problem, the study uses the average amount of bribery in a firm’s location, sector, and the country as an instrumental variable (IV). The results obtained are not consistent across country groups classified by national income. Due to obstacles during a firm’s operation, the amplitude of the positive effect of bribery on exports is reduced. Keywords: bribery; export; firm’s growth obstacles; National Income Classification; instrumental variable; probit regression 1. Introduction Bribery is a global problem and exists in all areas of economics, society, and politics. No country in the world is immune from corruption. Bribery can take many forms, such as bribing officials to secure contracts, or lobbying and kickbacks aimed at distorting institutional activities. This behavior can create inequality in society, and cost the economy. Thus, addressing all of them is crucial to achieving sustainable progress and change. One reason behind the increase in bribery is participation in international marketsOne reason behind increasing bribery is the reliance on international trade integration. Petrou and Thanos (2014) . Therefore, the bribery situation in developing countries and transition economies is more complicated than in developed countries Olken and Pande (2012). Nevertheless, business bribery appears most often, hidden behind various guises Luo (2005). This activity aims to benefit the firm by smoothing administrative procedures, receiving preferential treatment, avoiding inspections, and audits, etc. As a result many previous empirical studies revealed that this behavior has a significant effect on firms’ performance, for instance, on business strategy Spencer and Gomez (2011), innovation Xie et al. (2019); Krammer (2019), competitiveness Beck and Maher (1989) , and export activity Gao et al. (2010). Integration is becoming an indispensable trend throughout economies in the world, especially in developing countries. Developing exports according to a sustainable and reasonable growth model is seen as a solution to open the economy to take advantage of foreign capital and technology of developing countries. The effects of exports, however, depend on the characteristics of the national economic system. In some cases, as shown, for example by Gros (2013) and by Lucarelli et al. (2018), in the case of EMU countries, they Economies 2022,10, 28. https://doi.org/10.3390/economies10020028 https://www.mdpi.com/journal/economies Economies 2022,10, 28 2 of 26 can destabilize the macroeconomy. Therefore, research on factors that affect the export of firms is an interesting topic and has practical value. This topic not only attracts the interest of macroeconomic managers but also receives the attention of business managers. Nonetheless, the firm’s decision-making is a harmonious combination of many complex factors and characteristics Coase (1937) . For example, Hellman and Schankerman (2000) suggested that small firms are more likely to be asked for bribes and pay higher informal fees than large ones. Because they are less able to negotiate and meet official requests. Corruption and solicitation of bribes arise from poor social governance Rose-Ackerman (2005). Institutional gaps caused by cumbersome regulations, unstable democratic politics and overlapping management methods, etc., create opportunities for harassment for certain individuals and sectors in the economy. In addition, barriers in the growth process of enterprises include internal difficulties (such as regarding labor, management, financial, etc.), and external obstacles (such as unfair competition, the increase of competitors, etc.) are also factors affecting the bribery behavior as well as the exportability of enterprises Svensson (2003). One of the most recent theoretical studies is Cariolle’s paper Cariolle and Sekeris (2021) , which proposes a theory explaining the export boom mechanism and promoting bribery. Her research paid attention to the reaction of exporters concerning bribery when there is a boom or bust in the export market. One conclusion of the study found that firms would use bribery to increase their market value. However, the variability of exports and bribery depends not only on the export market size but also on the firm’s level of human capital. If enterprises have low human capital, they will seek to increase bribery to expand the market. Then, a positive export shock will reduce the bribery balance. By contrast, if the enterprise has a high level of human capital, the export market value, and a large export market size, the enterprise still tends to bribe a lot to gain market share. A positive export shock will reduce the bribe balance when there is weak revenue complementarity between market value and output. As an effect, the interaction between export and bribery needs to be considered in terms of the moderation of firm’s growth obstacles such as institutions, internal, and external barriers. Recognizing the importance of exports as well as bribery in enterprises, this study focuses on addressing research questions as follows: (1) How do a firm’s growth obstacles affect a firm’s bribery payment? (2) How do a firm’s bribery payments affect exporting under the operation obstacles? (3) How do obstacles moderate the relationship between bribery and export? To solve these questions, the study uses the cross-country data of the World Bank to focus on two primary analyses: (i) investigate the correlation between a firm’s growth obstacles and bribery payments; (ii) estimates a relationship between bribery and a firm’s activities, particularly export activities. Subsequently, by controlling for additional moderating variables, the study demonstrates the interaction of barriers with the influence of bribe payments on a firm’s export. The instrumental variable (IV) approach is applied to address endogenous problems. Moreover, since the model variables are binary and proportional, the models are estimated by Probit regression with instrumental variabe (IV-Probit regression). While some empirical studies found a positive effect of corruption on a firm’s output and labor capacity in Indonesia (an Asian country) Mendoza et al. (2015); Vial and Hanoteau (2010), some experts found the opposite effect in African factories Mcarthur and Teal (2002); Fisman and Svensso (2007). These conclusions imply that the “greasing” or “sanding" of the wheels depends on exogenous factors such as country characteristics (for example national corruption rate CPI, gross domestic product (GDP)). Therefore, by classifying data groups according to several different criteria such as country income, the export status of firms, and firm size, this study contributes empirical results to the literature on the linkage between export and bribery. The study provides cross-group comparisons to highlight the extent of bribery and barriers on firms’ exports under specific conditions of firm and country characteristics. These results can have many implications for policy makers and Economies 2022,10, 28 3 of 26 business managers. Besides, the study performs further analysis by adopting different clusters by firm size. In the remainder of the paper, the literature review is described in detail in Section 2. Section 3offers the baseline empirical specification and data description. Next, all results are demonstrated in Section 4. Furthermore, finally, Section 5is a conclusion. 2. Literature Review This section reviews some typical modeling and optimization work to understand and collect necessary experiences for the coming steps. 2.1. Review of Theoretical Models Rose-Ackerman (1975) was a pioneer in research on the “economy of corruption”. Rose’s corruption model showed a link between market structure and bribery, with a focus on policy elements that prevent corruption. She stated that bribery is quite common in the private sector. Based on the Rose’s model, Svensson (2003) found evidence of a positive relationship between bribery rates and the authorities’ control over firms’ activities. This result implied that obstacles in a firm’s operations increase its likelihood and the cost of bribery. Besides, he pointed out the probability that firms offer bribes. Enterprises with a high bribery potential are the ones with more activities involving the authorities because this creates more opportunities for harassment, such as firms engaged in import and export, investment loans, restructuring, etc. Furthermore, their study discussed an interesting and rather important conclusion that officials solicit bribes based on the firm’s ability to pay. In other words, the firm’s characteristics such as current profit, expected profit, etc. will affect the amount of bribe. However, in contrast, each enterprise can only pay bribes to a limited extent depending on its characteristics. A common approach to bribery is the queuing theory supported by Cobham (1954); Wishart (1960); Kleinrock (1967); Lui (1985). This theory assumed that bribery depends on the customer’s ability to pay and the satisfaction level. In particular, the client’s bribe amount depends on their ability to pay, expected position in the queue, and their taste for waiting. These findings showed that customer waiting time costs appear to fall with average bribery rates. However, as bribes increase, some weights in the bribery function model will change, causing this cost to grow again, although the rate of increase is slower than the initial decrease rate. Unlike previous bribery models often referred to as static models, Wu and Lan (2018) built dynamic models of firm-level bribery decisions to determine their bribe’s size and optimize its future value. The author considered the firm’s status concerning bribery in the last term. Additionally, Wu examined the role of infrastructure obstacles in the bribery-production nexus. Moreover, in Wu’s study, bribery is intended to lubricate the business wheels either actively or passively. Therefore, bribery is seen as a business strategy and not audited like bribing government officials. In other words, the author ignored the illegality of bribery. One exciting point is that he built a firm-level model to explain the bribery decisions of firms and developed aggregate-level analysis for many firms in industries. Additionally, Wu provided suggestions for applying the bribery control policy after analyzing decision-making processes at both the firm and industry levels. Accordingly, a strategy that increases the cost of bribery, such as strictly monitoring firms’ financial activities or eliminating the post-bribery benefits, can be applied, such as removing lower infrastructure impediments obtained by paying bribes. According to the author, both of these policies can successfully reduce the number of companies engaging in bribery by 50%. Furthermore, there are also other notable theoretical frameworks for bribery. For example, the equilibrium model that built on the social balance between the agencies’ effort to optimize the bribe’s revenue and the goal of optimizing a firm’s profit Henderson and Kuncoro (2004); Kaufmann and Wei (1999); the games theory analysis Henderson and Kuncoro (2004); Macrae (1982); the bargaining model between bribe givers and public servants Ryvkin and Serra (2012); and a principal-agent model Groenendijk (1997). Economies 2022,10, 28 4 of 26 Applying background models, the collection of empirical studies on bribery is vibrant. Within the study’s limits, the next section of literature presents bribery concerning commerce. 2.2. Review of Empirical Studies: “Greasing the Wheels” versus “Sanding the Wheels” The empirical literature on bribery shows that its impact on economic performance and firm performance is inconsistent. A series of documents indicate that bribes to state officials can help “greasing the wheels” of commerce without undermining the competitiveness of businesses. Meanwhile, many scholars believe that corruption harms firms, as “sanding the wheels”, ultimately to the detriment of economic development. Consequently, two empirical research groups show conflicting evidence on bribery effects as follows: Many researchers found shreds of evidence showing that bribe payments bring enterprises certain advantages. Proponents of “corruption is effective” believe that bribery is an incentive for firms to overcome regulation and conduct business more efficiently Leff (1964) ;Huntingtion (1970). Because bribery encourages authorities to speed up licensing, shorten waiting times for government services, and improve public services’ quality Kleinrock (1967) ;Lui (1985). Simultaneously, bribery created competition among government officials, leading to a significant improvement in the quality of governance Leff (1964). To support this view, Mendoza’s survey of over 2000 SMEs in 30 Philippine cities found evidence that bribery helps companies “smooth” their operations effectively. In particular, the author found no clues about the growth inhibition or performance reductions of firms caused by bribery. Furthermore, several studies using data from high-corruption countries showed that bribery facilitates enterprises to overcome institutional barriers, ineffective public services and improves firm performance Hellman and Schankerman (2000); Dreher and Gassebner (2013). Furthermore, bribery allows them to access valuable business opportunities and enjoy preferential treatment by building relationships with the government Gamage (2019). As a result, firms increase their ability to participate in the international market as their position in this market is increasingly enhanced Meon and Weill (2008). Additionally, empirical results in Chinese firms showed that policy uncertainty significantly affects bribery as a driver of corporate product innovation Xie et al. (2019). Research results found a positive relationship between innovation and bribery, supporting the view that bribery payments enable firms to overcome administrative obstacles in emerging countries Krammer (2019). A first-of-its-kind study on the interaction between corruption and product innovation by firms in Vietnam confirmed wheel lubrication. Nguyen et al. (2016) explained that innovation can be seen as a short-term transaction to introduce a product in Vietnamese SMEs. Thus, informal payments (as a bribe) can be effective in these transactions and reduce these processes’ risks and difficulties. These results were similar to those of Krastanova (2014). He claimed that Bulgarian companies also receive significant and positive effects from bribery in their innovation activities. Nevertheless, Mahagaonkar’s research Mahagaonkar (2009) has obtained exciting results on the relationship between corruption and innovation in African countries. This research revealed that both exist simultaneously “greasing the wheels” and “sanding the wheels” when the author categorized innovation into four distinct categories: process, product, organization, and marketing. The obtained results indicated that bribery has only positive effects on marketing innovation and an adverse impact on organizational and product innovation. In addition, the author found no evidence that bribery affects process innovation. Shleifer and Vishny (1993) also acknowledged that innovation improves when firms engage in bribery, and yet an enterprise that does bribe is easily distrusted. For this reason, Shleifer suspected that these positive numbers are the result of fraud at the bribery firms. These firms dishonestly reported on the state of innovation of their firm. Thanh et al. (2021) found a positive relationship between bribery and exports. Exporters tend to spend more on bribes. This state exists because the export market carries more risks and requires more standard requirements than the domestic market. Moreover, entering the export market requires enterprises with high enough labor productivity that can bear the export sunk costs Bernard et al. (1995); Aw et al. (2000). Economies 2022,10, 28 5 of 26 On the other hand, exploring a second aspect of the matter, many empirical publications have proven that bribery aggravates and inhibits the development of enterprises in many regards. Firstly, bribery makes firms tend to increasingly rely on illegal actions to gain benefits, instead of focusing on improving production quality and productivity. Thus, increased bribery leads to a decrease in average labor productivity over time De Rosa et al. (2010); Dal Bo and Rossi (2007) . Supporting this idea, Campos et al. (2010) believed that bribery firms tend to offer bribes more seriously in the future because the probability of continuing to make a bribe in the next period is substantial, with increasing size. Additionally, bribery is also a solid barrier to new entrants Birhanu et al. (2016). Secondly, bribery increases the production and financial costs of enterprises Kaufmann and Wei (1999); Wu and Lan (2018). As a consequence, it reduces the profit from business activities and consequently reduces financial resources’ investment and market expansion capacity Birhanu et al. (2016). Investment growth by firms tends to decrease as bribe payments increase, especially in transition economies Asiedu and Freema (2009). In addition, bribery puts pressure on firm’s innovation Ayyagari et al. (2014). In particular, he studied 25,000 companies in 57 countries and found that those with product innovation had 0.37 higher bribe costs than those that did not. Then, Goedhuys et al. (2016) also found similar results for firms in Egypt and Tunisia. Additionally, a recent study based on Vietnamese SMEs’ 2015 data also supports the hypothesis that firms using bribes as a business strategy will reduce motivation and hinder innovation in a company Nguyen (2020b). Thirdly, bribery might bring some short-term advantages such as helping firms to pass product quality checking or to break the standard rule about product safety, etc., are also pointed to in other papers Gamage (2019); Svensson (2003); Meon and Weill (2008). As an effect, firms can sell goods of inferior quality or sell goods earlier than competitors. Albeit with short advantages, bribery causes losses for the goods market, consumers, and firms in the long term, such as loss of reputation, loyal customers, and long-term profits Nguyen et al. (2016); Rand and Tarp (2012). Furthermore, bribery did not have the expected effects on some other firms’ activities. Wellalage et al. (2020) illustrated that bribery SMEs in India face 68% higher credit constraints than non-bribery counterparts. At the same time, a firm that engages in bribery increases the likelihood of its credit constraints by more than 30%. Seker and Yang (2012); Nguyen (2020a) demonstrated that bribery can affect the revenue growth of firms regarding different firm sizes. The revenue of bribery firms tended to be 0.12 points lower than that of non-bribery firms. In addition, by classifying the purpose of bribery, Nguyen found a critical conclusion that some bribery for one purpose may be more beneficial than another. However, no evidence was found that targeted bribery makes a difference in labor productivity. In other words, overall, the net impact of bribery on firms is still negative. Last but not at least, some “greasing the wheels” scientists argued that a firm’s solid and favorable position in the domestic market creates little incentive to encourage exports Ito and Pucik (1993) ;Hundley and Jacobson (1998). Accordingly, bribery creates advantages for firms in the domestic market cause of reducing the size of exports (Cuervo- Cazurra (2006)). These advantages make enterprises lose motivation to find customers or export goods to foreign markets, where they will no longer receive benefits as in the domestic market. Supporting this view, Lee and Weng (2013) found a negative relationship between bribes and firm export intensity. Specifically, a 1% increase in corporate bribery will reduce the power of exports to 1.43%. A similar result across 25 countries, mainly including countries in Eastern Europe and Central Asia, was found when Gamage (2019) analyzed the relationship between bribery payments and export intensity and found that the bribe payment rate increased from 0% to 8%, the firms’ export intensity decreasing by 9.25%. However, according to Olney’s findings Olney (2016) , the effect of bribery on exports was only fully discernible if direct and indirect exports are classified separately. The results provided new evidence that the firms’ ability to export indirectly is increased if the firm makes a bribe. On the contrary, it reduces the ability of companies to export directly. Thereby, the author made an important suggestion in this study for developing Economies 2022,10, 28 6 of 26 countries where corruption is common. The intermediaries play an essential role in making a profit for the trade in making a profit for trade. Firms in these countries can enjoy success in exporting indirectly through intermediaries. In summary, it can be seen that this interesting relationship is highly dependent on factors such as the characteristics of the firm, country, and survey period, etc. In particular, placed in different environments, this effect may be slightly deflected or reversed. Therefore, the literature on the influence of obstacles on bribery will be considered in the following section. 2.3. The Regulatory Role of Obstacles The political instability creates a weak and unstable business environment Hiatt and Sine (2014). This instability hinders the development and causes adverse effects on the benefits of firms Garcia et al. (2008). In addition, it increases dependence on the government, one of the reasons for increasing corruption and bribery Xie et al. (2019). The strand of bribery literature revealed that regardless of the bribery form, “grease” or “sanding” the wheel, bribery depends a lot on the institutions and barriers to a firm’s operation, such as financial capacity, competitiveness, and a network of enterprises. A weak institutional system is a cause of limiting the export capacity of enterprises. This fact is due to enterprises having difficulty obtaining export licenses and facing too many other complicated regulations Rose-Ackerman (1975). Krammer (2019) researched bribery in 30 emerging markets in Central Asia and Eastern Europe and showed that bribery offers many benefits to firms looking to innovate. However, whether this effect is positive or not depends on the formal and informal pressures of the institutional environment. He argued that bribery will be more successful in facilitating the launch of new products in countries with weak standard regulatory controls over corruption. These environments provided opportunities for officials to extort bribes. In turn, the already constrained institutional roadmaps become smooth, allowing companies to deploy new products more quickly Luo (2005). In contrast, solid formal institutions limited the influence of bribery on new product launches. Additionaly, Svensson (2003) stated that the higher the barriers a firm faces, the more likely a firm will pay bribes. If a firm does not pay for this, it would have to exit the industry. Otherwise, a firm can refuse to pay bribes when obstacles are low without worrying about any retaliation. The economy has many constraints on market entry that will hinder new businesses. Dreher and Gassebner (2013) found clear evidence that the number of new entrants is decreased when entry regulations are too complicated. This negative effect is reduced once the enterprise makes a bribe. However, Dreher’s claims were only considered short-term. The author doubted that bribery, or the influence of strict regulations and bulky legal procedures, is unknown in the long run. Furthermore, Vial and Hanoteau (2010) argued that institutions influence economic activity by affecting exchange and production costs. Their study highlighted the threshold effects of institutions on the relationship between corruption and economic growth. Three institutional characteristics observed in the survey are political stability, property rights, and the political system. Political stability boosts production. When politics is stable, corruption and bribery cause disadvantages for tricky politics. However, in cases of severe political instability, corruption can help tie the economic system together for a while. Notwithstanding, Klapper et al. (2006) only found the negative effect of institutional and regulatory barriers on the market entry opportunities of new firms in countries with low corruption rates or developing countries. For developing countries or high corruption index countries, the impediment of regulatory barriers does not cause a decrease in the number of new entrants to the market. The reason is that the bribery of these businesses can smooth the obstacles. In summary, the reviewed literature demonstrated a comprehensive analysis of bribery and firms’ activities, including exports. Some of them analyzed the impact of institutions Economies 2022,10, 28 7 of 26 on bribery on some firms’ aspects, such as innovation and market entry. Therefore, this study is expected to connect and clarify the influence of firm’s growth obstacles in the interaction of bribery and export in separate analysis groups. 3. Research Approach and Data Description This section focuses on two sets of main analyses. The first one scrutinizes the integration of obstacles and bribery payment. The second one analyses the effect of bribery on export intensity under a firm’s operation obstacles. Models are performed in turn on four separate income country groups, including low-income countries (LI), low-medium income countries (LMI), upper-medium income countries (UMI), and high-income countries (HI). 3.1. Estimation Strategy Analysis 1—Influence of Obstacles on Bribery Based on their characteristics, firms offer a suitable amount of bribes to lubricate the wheels, thereby reducing the pressures of barriers in their operation. Consequently, a simple regression model is proposed to test the negative relationship between the two variables bribery and obstacles as follows: BriSijs =α1+δ1·Obsijs +vs.+γ+eijs (Model 1) (1) BriSijs =α2+δ2·Obsijs +σ2·Controlijs +vs.+γ+eijs (Model 2) (2) Brisijs represent bribery share of firm i, Obsijs is the set of obstacles, while Controlijs is the set of firm characteristics as a control variable. eijs is an error term. All symbols and subscripts are maintained throughout this study. Previous growth economic theories of firms have confirmed that an enterprise operating in the economy, of course, always has many binding relationships Box (2008). On the top of firms’ characteristics, environmental and structural factors also play an essential role in the success or failure of the company. Many scholars have classified firms’ obstacles into several categories: internal organizational obstacles, obstacles due to external market position, institutional obstacles, and financial obstacles Bartlett and Bukviˇc (2001). In particular, internal barriers include the specific characteristic of the enterprise such as shortage of skilled workers, weak management capacity, etc. These factors are considered the foundation of a firm’s operations, thus determining its success or failure Watson et al. (1998). Likewise, the barriers regarding competition, the general situation of a firm in the industry, demand for the product, access to raw materials, commercial rules, etc. are considered obstacle factors from the market environment outside the company. These factors can support or hinder the operation and survival of the business Olawale and Garwe (2010) . One of the most challenging barriers worthy of attention in enterprises are financial barriers, including lack of equity capital or barriers concerning credit access such as high credit costs, bank fees, and collateral constraints. Researchers point out financial constraints as playing an essential role in the business and investment decisions of enterprises. For example, Kapplan and Zingales (1997) found the sensitivity of financial constraints to firms’ investment cash flows; Greenaway et al. (2007) ;Bellone et al. (2010) showed a link between financial restrictions and decisions export. Finally, institutional obstacles play a critical role in firm performance. According to North (1989), institutional constraints include norms, conduct practices, and judicial rules that a company encounters. Thus, these difficulties at different levels affect the development of enterprises. Although many firms expect to solve problems with bribes, bribery is indeed unlikely to solve all problems. Gamage (2019) showed that several barriers interfere directly with a company’s bribery decision, but others interfere indirectly. For example, the tax bracket applicable to a business is determined by law, which is very difficult to change by bribery or the pressure from competitors has a binding relationship with the company’s outcome. These factors may not be the factors that directly influence the firm’s bribery decision but are the indirect factors that motivate the enterprise to bribe to expand production. Economies 2022,10, 28 8 of 26 As a consequence, in these models, I choose a set of three obstacles including tax administration (marked as taxad), business licensing and permit constraints (denoted as permit), and political instability (denoted as pol) as direct factors that affect bribery of a firm. Thus, the set of obstacles is Obsijs ={taxad,permit,pol}. In addition, because model 1 includes a fractional dependent variable, the outcomes observed are bounded within the range of [0, 1]. According to Wooldridge (2015), the Probit regression is suitable. Then, I report the predictive margins of each obstacle on firm bribery payment over a firm’s export status. Thus, the results provide a detailed and comprehensive view of the impact of each perceived impediment on bribery at 95% confidence intervals, comparing exporting and non-exporting firms. Analysis 2—Influence of Bribery Payment on Export Intensity To shed light the impact of bribery on export intensity which includes all relevant control variables, I regress model (3)1: ExSijs =α3+β3·BriSijs +σ3·Controlijs +vs.+γ+eijs (Model 3) (3) where subscripts i , j , s denote firm, country, and sector, respectively. v , γ are year and sector fixed effect. Brisijs represent bribery share of firm i , Obsijs is the set of obstacles, while Controlijs the set of firm characteristics as a control variable. eijs is an error term. Then I scrutinise the influence of paying bribes on the export percentage under a firm’s growth barriers by adding some obstacles into Model 4. As mentioned above, a firm’s growth obstacles might be divided into four groups. Instead of using institutional obstacles as analysis 1, this section selects obstacles that are likely to affect business operations (such as exports) and indirectly affect bribery in the three remaining groups. For simplification, I select one obstacle from each of the remaining groups. The obstacle representing each group is the one that most businesses consider to be the biggest obstacle. They are practices of a competitor in the informal sector (denoted as compe), inadequacy of a skilled worker (denoted as inade), and financial constraint (marked as fin). ExSijs =α4+β4·BriSijs +δ4·Obsijs +σ4·Controlijs +vs.+γ+eijs (Model 4) (4) where Obsijs =compe,inade,fin is the set of obstacles. Because ExSijs is a fractional dependent variable, I apply the Probit method to estimate Equation (models 3 and 4). In addition, to eliminate the unobservable factor specific to countries and sectors, I use v , γ as year and sector fixed effect Thanh et al. (2021). In addition, as an endogenous issue of bribery, models (3) and (4) are then estimated by using Instrumental Probit Regression (IV-Probit regression) with “the location-country-sector average of bribery” as an instrumental variable Bernard et al. (1995), Thanh et al. (2021). A simple procedure for estimating this model is that the bribe variable is assessed as a linear equation of the instrumental variable and other explanatory variables to estimate bribes’ value in the first stage Wooldridge (2015) . This stage is expected to indicate a strong correlation between firm-level bribery and instrument variables, implying that the province-sector average bribery rate is a relevant tool for explainiing firm-level bribery. Then, export is estimated as a function of the values calculated from the first stage and other exogenous variables. In all regressions, I report the average margin effect. Further Sensitive Analysis (i) Consider Each Obstacle Separately Instead of including all three obstacles (competition, inadequately skilled worker, and financial obstacles) as model 4, in this section I examine model 4 for each barrier to comparing the effect of a firm facing and not facing an obstacle. Through this process, the interaction effect between obstacles and bribery payment can be observed more thoroughly. Notwithstanding that these constraints do not directly affect the bribery decision, their existence can cause pressure on the enterprise’s business process. From there, indirectly, Economies 2022,10, 28 15 of 26 a firm’s export as the share of export sales in LMI and UMI groups is likely to increase to 2.1% and 3.6% point once a firm increase bribe, respectively. By contrast, there is a negative relation between bribery payment and the firm’s export share in nations with a high-income level HI (column 7). Approximately 11.4% drop in the probability of exports was recorded at a 5% significant level ( p< 0.05) when a firm pays more for informal payments. These findings are similar to Gamage (2019) covering 25 countries in Eastern Europe and Central Asia. She found that the bribe payment rate increased from 0% to 8%, the firms export intensity decreasing by 9.25%. Nevertheless, there is no evidence for the effect of bribery on exports in the LI country group. The coefficient of bribery is not statistically significant in both models that exclude and include control variables (columns 1 and 2). In addition, comparing the results of model 3 and model 4, the influence of obstacles makes the correlation coefficient of bribery decrease, though the gap is negligible. This result shows that bribery is less effective when enterprises face obstacles such as an external competitive environment, lack of skilled labor, and higher financial constraints. Table 4. The effect of bribery payment on export in different country groups. LI LMI UMI HI (1) (2) (3) (4) (5) (6) (7) (8) briS 0.007 0.006 0.021 *** 0.019 *** 0.036 *** 0.035 *** −0.114 ** −0.086 * (0.016) (0.016) (0.007) (0.007) (0.005) (0.006) (0.050) (0.052) compe −0.092 ** 0.010 −0.040 *** −0.132 *** (0.038) (0.011) (0.011) (0.018) inade 0.083 * 0.083 *** 0.021 * −0.002 (0.044) (0.011) (0.012) (0.017) fin 0.091 ** −0.029 *** 0.022 * 0.007 (0.038) (0.011) (0.012) (0.018) fage 0.023 0.022 −0.003 −0.005 0.081 *** 0.083 *** 0.074 ** 0.074 ** (0.064) (0.065) (0.018) (0.018) (0.020) (0.020) (0.032) (0.032) fsize 0.323 *** 0.330 *** 0.346 *** 0.346 *** 0.259 *** 0.256 *** 0.308 *** 0.302 *** (0.038) (0.039) (0.010) (0.010) (0.011) (0.011) (0.019) (0.019) mae 0.002 0.004 0.165 *** 0.162 *** 0.091 *** 0.092 *** 0.038 0.038 (0.067) (0.067) (0.020) (0.020) (0.021) (0.021) (0.033) (0.033) RnD 0.213 * 0.183 0.281 *** 0.274 *** 0.225 *** 0.225 *** 0.469 *** 0.473 *** (0.122) (0.124) (0.033) (0.033) (0.034) (0.034) (0.054) (0.054) ctfc 0.413 *** 0.390 *** 0.383 *** 0.385 *** 0.432 *** 0.432 *** 0.311 *** 0.291 *** (0.118) (0.119) (0.029) (0.029) (0.031) (0.031) (0.045) (0.046) cau −0.011 −0.031 −0.043 −0.042 −0.056 * −0.055 * −0.005 −0.021 (0.085) (0.085) (0.031) (0.031) (0.030) (0.030) (0.056) (0.056) gos −0.064 −0.055 −0.070 ** −0.073 ** −0.072** -0.077 ** −0.054 −0.041 (0.109) (0.110) (0.035) (0.035) (0.036) (0.036) (0.061) (0.061) own 0.646 *** 0.667 *** 0.689 *** 0.682 *** 0.573 *** 0.571 *** 0.601 *** 0.586 *** (0.107) (0.109) (0.043) (0.043) (0.047) (0.047) (0.067) (0.068) inno −0.007 −0.016 0.104 *** 0.087 *** 0.109 *** 0.113 *** 0.160 *** 0.189 *** (0.110) (0.111) (0.031) (0.032) (0.032) (0.032) (0.049) (0.050) Const −1.744 *** −1.818 *** −2.769 *** −2.862 *** −2.094 *** −2.115 *** −1.455 *** −1.294 *** (0.532) (0.541) (0.224) (0.226) (0.212) (0.215) (0.359) (0.360) Obs 1.516 1.516 16.228 16.228 11.787 11.787 4.549 4.548 Note: All regressions use the location-sector -country average of bribery as the instrumental variable. Standard errors in parentheses. ***, **, * denote significance at 1%, 5%, 10%. All regressions include the year and sector-fixed effects. Columns 1, 3, 5, and 7 are the results of regressions without the set of obstacles (model 3). Columns 2, 4, 6, and 8 are the results of regression with the set of obstacles (model 4). LI, LMI, UMI, and HI are four country groups regarding the national income. Economies 2022,10, 28 16 of 26 A part from this, in the set of control variables shown in the result table, firm size, international certificate, RnD, and foreign ownership are factors that have a notable effect on export activity. All characteristics positively affect exports and are statistically significant at the 1% level in all regressions in which the large-sized firm is more likely to export than smaller firms (30% on average). Similarly, the higher the foreign ownership and/or holding the international certification, the more likely the firm increases exports (40% and 60% on average, respectively). Research and development activity tends to powerfully impact the export probabilities of firms in a developed economy (HI group). RnD can increase the likelihood of export growth of this group by nearly 47% (column 8), nearly double that of the other groups. In addition, there is no evidence for the impact of state contracts and capacity utilization on export activity in these estimates. I predict the margin effect of bribery payments on a firm’s export share to comprehensively analyse results. Through the intriguing results of marginal effect ranges, it is possible to analyze the responses of exports to changes in each level of bribery. Table 5presents the detailed results. In the LI group, starting from over 50% of bribery payments, the margin effects are no longer significant, as ( p> 0.1). When a firm pays more bribery from 0% to 40%, the probability of export increases by 27.9% (column 1). Similarly, the results in LMI are significant up to the 30% rage of bribery payment (Column 3). In addition, the results in the UMI group are statistically significant in most of the bribery ranges, except the 20% range (Column 5). Compared with the same degree of bribery payments, once a firm raises a bribery payment from 0% to 30%, export probability also goes up in both LMI and UMI groups. Export’s likelihood in UMI countries increases by approximately 106%, over 1.5 times higher than in the LMI group. Finally, the more extensive the bribery range, the lower the likelihood of exporting firms in the HI group. The results in the HI group are significant across all ranges of bribery. These results infer that when the bribe value increases by 30%, the probability of exporting decreases from probability ( − 0.107) to probability (−2.658). Table 5. Predictive margins of bribery share on export. LI LMI UMI HI Bribery Share Margin Std. Margin Std. Margin Std. Margin Std. (1) (2) (3) (4) (5) (6) (7) (8) 0% −1.294*** 0.059 −1.102 *** 0.015 −0.699 *** 0.014 −0.107 *** 0.025 10% −1.224 *** 0.132 −0.897 *** 0.067 −0.346 *** 0.056 −0.966 ** 0.486 20% −1.154 *** 0.277 −0.692 *** 0.135 0.006 0.112 −1.825 ** 0.985 30% −1.084 ** 0.426 −0.486 ** 0.203 0.358 ** 0.167 −2.685 ** 1.484 40% −1.015 * 0.575 −0.281 0.272 0.711 *** 0.223 −3.544 ** 1.983 50% −0.945 0.726 −0.075 0.340 1.064 *** 0.280 −4.403 ** 2.482 60% −0.875 0.876 0.130 0.409 1.417 *** 0.336 −5.263 ** 2.981 70% −0.805 1.026 0.335 0.478 1.770 *** 0.392 −6.122 ** 3.481 80% −0.736 1.177 0.541 0.547 2.122 *** 0.448 −6.981 ** 3.980 90% −0.666 1.327 0.746 0.615 2.475 *** 0.504 −7.841 ** 4.479 100% −0.596 1.478 0.952 0.684 2.828 *** 0.560 −8.700 ** 4.978 Obs 1520 16,228 11,787 4549 Note: *** p< 0.01, ** p< 0.05, * p< 0.1. Std. represents standard error. In a nutshell, this study finds the positive effect of bribery payment on firms’ export in nations belonging to the medium-income countries group, but negative influence in highincome countries. Moreover, no evidence is found for the rest group (LI). Moreover, the effect of bribery on exports becomes more pronounced when considering firms’ obstacles. Economies 2022,10, 28 17 of 26 The visualization of the margin results provides a clearer view of the impact of perceived barriers on the relationship between bribery and exports. Further Sensitive Analysis Results (i) Consider Each Obstacle Separately In contrast to the main model, instead of controlling for the effects of all three constraints simultaneously, this section examines the interaction of each in turn on the relationship between bribe payment and exports. The similarity with the main part of the model is that the set of control variables remains the same, and the regressions are performed on 4 sub-data classified by country income group. Table 6reports the results of the regressions. From the comparison between firms without difficulty (No) and with difficulty (Yes), it is possible to highlight the interaction between obstacles and bribery in relationship with exports. Table 6. The effect of bribery on export under each obstacles separately. Competition Inadequate Skilled Worker Financial Constraint No Yes No Yes No Yes (1) (2) (3) (4) (5) (6) Panel A: LI briS −0.095 0.012 −0.029 0.009 −0.088 0.0073 (0.081) (0.016) (0.071) (0.016) (0.073) (0.017) Control variables YES YES YES YES YES YES Const −3.515 *** −1.255 * −2.634 *** −1.366 ** −2.862 *** −1.217 ** (1.104) (0.647) (0.916) (0.604) (1.221) (0.619) Obs 453 1049 548 941 376 1131 Panel B: LMI briS 0.038 *** 0.015 * 0.0192 * 0.018 ** 0.031 ** 0.021 ** (0.014) (0.008) (0.014) (0.008) (0.013) (0.008) Control variables YES YES YES YES YES YES Const −3.094 *** −2.708 *** −3.056 *** −2.702 *** −2.869 *** −2.793 *** (0.421) (0.269) (0.376) (0.282) (0.387) (0.276) Obs 5941 10,278 6636 9592 5362 10,866 Panel C: UMI briS 0.046 *** 0.037 *** 0.056 *** 0.033 *** 0.064 *** 0.033 *** (0.014) (0.006) (0.011) (0.006) (0.014) (0.006) Control variables YES YES YES YES YES YES Const −1.672 *** −2.332 *** −1.791 *** −2.292 *** −2.228 *** −1.980 *** (0.327) (0.284) (0.329) (0.286) (0.359) (0.265) Obs 4278 7509 4236 7551 4425 7362 Panel D: HI briS 0.036 −0.186 *** −0.007 −0.173 *** −0.131 −0.085 (0.09) (0.056) (0.075) (0.066) (0.097) (0.059) Control variables YES YES YES YES YES YES Const −1.500 *** −0.898 * −1.439 ** −1.346 *** −2.084 *** −1.112 ** (0.523) (0.537) (0.715) (0.431) (0.530) (0.522) Obs 1963 2580 1154 3395 1932 2608 Note: Standard errors in parentheses. ***, **, * denote significance at 1%, 5%, 10%. "NO" means a firm does not face a constraint. "YES" means a firm faces a constraint. All regressions include the year- and sector-fixed effects. The set of control variables is the same main analysis (include fage, fsize, mae, RnD, ctfc, gos, own, and inno). Economies 2022,10, 28 18 of 26 Overall, bribe payments have absolutely no relationship with exports in the LI group. Moreover, the presence of hindrance changed the relationship between bribery and export, from insignificant to significant in HI group. In the other two groups, the correlation coefficients are significant, regardless of whether the enterprise faces obstacles or not. Specifically, panel A reports that the rate of bribe payment of enterprises in the LI group is not able to explain the change in export probability, because the correlation coefficients are all significantly greater than 10%. This result is consistent with the results in the main model (Table 4). Comparing the number of observations between the groups facing and not facing obstacles, the number of enterprises with difficulties is more than that of enterprises without obstacles. Even the number of firms in LI that are financially constrained is three times more likely than unrestricted firms (columns 5 and 6). However, compared with the number of observations in the rest of the country groups, the figure in the LI group was the lowest, ranging from 376 to 1131 observations. The limitation of the sample may be the reason why we question the consistency of the population. Panel B and panel C reflect the positive outcomes of bribery in the LMI and UMI groups. Although the correlation coefficient is significant even when firms do not face impediments, the presence of impediments reduces the magnitude of the impact of bribery on exports, comparing columns 1, 3, 5 and columns 2, 4, 6, respectively, in pairs. These results are similar to the main part (results in Table 4), but the effect of bribery on export probability is stronger when observing each type of hindrance separately. To illustrate this point, for firms in LMI facing financial difficulty, the effect of bribery is strongest. The probability of exporting is then likely to increase by 0.021 (column 6), compared with 0.015 (column 2) and 0.018 (column 4) once these businesses face competition, or lack of skilled workers, respectively. In contrast, the effect of bribery was strongest for firms in UMI that face competitive difficulties (correlation coefficient 0.037), compared with firms facing the other problems. Interestingly, however, when comparing the disparity caused by the impediment, the results reflect an opposite trend. The most significant reduction in the impact of bribe payments in the LMI group (panel B) was due to competition, from 0.038 (column 1) to 0.015 (column 2). Whereas, in the UMI (panel C), the change caused by financial constraints is the most pronounced, from 0.064 (column 5) when a firm is not facing financial difficulties to 0.033 (column 6) when it is facing financial constraint. Finally, no different from the main result (Table 4), panel D reflects the negative outcome of bribe payments to exports by firms in the HI group. Under the interaction of competition and lack of skilled workers, the association between bribery and export becomes statistically significant. The magnitude of the effect of bribery when considering the interaction of bribery and these two constraints is − 0.186 (Column 3) and − 0.173 (Column 4), respectively. This result is much higher than the effect of bribery when controlling for all three constraints in the main model simultaneously (Column 8– Table 4 ). The cause of this finding is that financial constraints do not change the nature of the relationship between bribery and exports in the HI group. The ability to export is not affected by bribery regardless of the firm’s financial position, as the correlation coefficients are not statistically significant (columns 5 and 6—panel D–Table 6). To sum up, the results obtained when separating each constraint are in favor of the main results. In more detail, the interaction between competition and bribery makes the effect of bribery on exports strongest in the UMI and HI groups. While the interaction between financial constraints and bribery makes the effect of bribery on exports strongest in the LMI group. (ii) SMEs and Large-Sized firms In this further analysis, the study approaches the main models on SMEs and largesized firms. First, the results of model 1 and model 2 for SMEs and large-sized firms are shown in Table 7. Regardless of firms’ size, the scale of a firm’s bribery tends to increase once firms face more severe difficulties in tax administration, business licensing procedures, and political instability (columns 1 and 3). This result may reduce when controlling for firm characteristics as control variables (columns 2 and 4). The results also show that Economies 2022,10, 28 19 of 26 political instability has the least influence on bribery payment compared to the other two barriers, regardless of SMEs or large companies. These findings aline with the results of the main part. Table 7. Robustness test results of analysis 1 by firm size. SMEs Large-Sized Firms Model 1 Model 2 Model 1 Model 2 (1) (2) (3) (4) taxad 0.110 *** 0.088 *** 0.105 *** 0.079 *** (0.005) (0.010) (0.012) (0.018) permit 0.103 *** 0.103 *** 0.137 *** 0.114 *** (0.005) (0.011) (0.012) (0.019) pol 0.054 *** 0.021 ** 0.080 *** 0.060 *** (0.004) (0.008) (0.010) (0.015) exp 0.000 −0.000 (0.000) (0.000) fage 0.045 *** −0.005 (0.016) (0.028) fsize −0.002 0.053 ** (0.013) (0.024) mae −0.112 *** 0.008 (0.016) (0.029) RnD −0.006 0.044 (0.030) (0.046) ctfc −0.163 *** −0.295 *** (0.028) (0.042) inno −0.005 *** −0.007 *** (0.000) (0.001) cau 0.308 *** 0.170 *** (0.027) (0.046) gos 0.099 ** 0.093 * (0.046) (0.050) own 0.162 *** 0.188 *** (0.025) (0.046) Const −2.595 *** −1.832 *** −2.837 *** −0.704 (0.06) (0.564) (0.171) (1.194) Obs 33,906 28,043 10,790 9547 Note: Standard errors in parentheses. ***, **, * denote significance at 1%, 5%, 10%. All regressions include the year and sector-fixed effects. In addition, Figure 2visually depicts the predicted marginal effects of these constraints in both firm-size groups at the 95% confidence intervals relative to the export situation. All three obstacles have a significant marginal impact on the payment of bribes at all obstacle levels, as p= 0.000. Overall, it is immediately apparent that the shapes of the graphs are similar between exporting and non-exporting firms in both SMEs and large firms. The correlation between tax impediment and the marginal prediction of bribery increases gradually with a positive linear function. However, the growth rate in the group of SMEs is slightly higher. Likewise, the graph of the political instability variable can also be predicted as an increasing linear function. While political instability has almost no discernible effect Economies 2022,10, 28 20 of 26 on the bribery behavior of SMEs, large firms find a notable variation in the probability of bribery when it faces with uncertainty politics. Illustrating this point, the graph is steep as the firm moves from the starting point (no obstacle) to the beginning of perceived difficulty due to corruption (minor obstacle). However, the degree of variation in the likelihood of bribery increases slowly at the next difficulty levels. Unlike the above two hindrances, the relationship between permit and the marginal effect of bribery is most appropriate with a saturation function, as the coefficients R2 are greater than 64% in SMEs, and over 98% in large firms. When a firm begins to perceive business licensing and permit impediments, the probability of bribery gradually increases in both groups of firms, before becoming saturated when impediment reaches Major level. In short, tax administration barriers have a stronger impact on bribery behavior of SMEs than large firms. In contrast, the latter is more dominated by political instability and bureaucratic difficulties. Figure 2. Visually depicts the predicted marginal effects of these constraints in both firm-size groups at the 95% confidence intervals relative to the export situation. Comparisons between non-exporters and exporters were made within each country group. In particular, subfigure 1 and 2 are in SMEs group, and subfigure 3 and 4 are in large firms group. Second, the effect of bribery on exports is tested by IV-probit estimation using the location–country–sector average of bribery as the instrumental variable (models 3 and 4). The results are presented in Table 8. Columns 1 and 2 reflect the results for SMEs. The results emphasize that an increase in bribe payments can account for an increase in exports (the coefficients are statistically significant at the 1% level, p< 0.01). SMEs paying more bribes can increase their ability to export. Specifically, in the baseline model (Model 3) and the model that controls for the specific characteristics of the enterprise (Model 4), the ability to export can increase to 1.6% and 1.8%, respectively. On the other hand, no relationship was found between bribe payments and exports in large firms (Column 4). The correlation coefficient between bribery and exports is not statistically significant even when I add control variables to the model (column 5). In addition, although competitive pressure from competitors reduces the export ability of both SMEs and large enterprises, Economies 2022,10, 28 21 of 26 its impact on SMEs is almost twice as high, − 0.074 (Column 2) compared with − 0.037 (Column 4). This finding is the same expected result as Ito and Pucik (1993). Furthermore, though the shortage of high-skilled labor is significant for export activities of enterprises, the regressions do not find a significant difference of this factor for the probability of paying bribes of SMEs and large corporations. Unlike the above two types of impediments, the study only found evidence of a negative interaction of financial constraints on the relationship between bribery and exports in the large group of firms (column 4). In contrast, no relationship of financial problems was found in the SMEs group (column 2). Table 8. Robustness test results of analysis 2 by firm size. SMEs Large-Sized Firms Model 3 Model 4 Model 3 Model 4 (1) (2) (3) (4) briS 0.016 *** 0.018 *** 0.008 0.007 (0.006) (0.005) (0.005) (0.005) compe −0.074 *** −0.037 * (0.008) (0.012) inade 0.065 *** 0.064 *** (0.008) (0.012) fin −0.012 −0.026 ** (0.008) (0.013) fage 0.038 *** 0.035 ** 0.033 *** 0.037 * (0.014) (0.014) (0.019) (0.020) fsize 0.336 *** 0.351 *** 0.185 *** 0.180 *** (0.012) (0.012) (0.017) (0.018) mae 0.091 *** 0.096 *** 0.132 *** 0.134 *** (0.015) (0.015) (0.020) (0.02) RnD 0.312 *** 0.308 *** 0.297 *** 0.300 *** (0.024) (0.024) (0.032) (0.032) ctfc 0.468 *** 0.463 *** 0.468 *** 0.461 *** (0.023) (0.022) (0.030) (0.029) inno 0.105 *** 0.104 *** 0.093 ** 0.085 ** (0.023) (0.024) (0.033) (0.032) cau −0.012 −0.021 0.013 0.001 (0.022) (0.023) (0.036) (0.036) gos −0.100 *** −0.103 *** −0.042 −0.041 (0.028) (0.028) (0.035) (0.035) own 0.681 *** 0.671 *** 0.567 *** 0.557 *** (0.037) (0.038) (0.038) (0.038) Const −2.455 *** −2.425 *** −2.291 *** −2.313 *** (0.120) (0.121) (0.220) (0.221) Obs 28.275 28.275 9.252 9252 Note: Standard errors in parentheses. ***, **, * denote significance at 1%, 5%, 10%. All regressions include the year and sector-fixed effects. All in all, regardless of firm size, obstacles related to tax administration, business licensing and permit, and corruption all explain the variation in corporate bribery probabilities. The relationship was found to be positive. However, the study only found a link Economies 2022,10, 28 22 of 26 between bribery and exports in the case of SMEs, but did not find any indication of this association in large firms. 5. Discussion and Conclusions An enterprise operating in production and business always faces many barriers regarding external obstacles (such as institutional, regulatory, and competitive difficulties of the business environment) and internal obstacles (such as the lack of capital and skilled labor, etc). On top of that, firms also always face negative pressures caused by harassment in various forms by administrative agencies such as lobbying, kickback, grease payment, etc. This situation seems to be more acute in countries with high levels of corruption, where there are gaps in government regulation to create space for fraud and solicitation. These barriers, directly or indirectly, affect the outcome of the firm. This paper adopts firm-level data cross-country provided by the World Bank, combined with the World Bank’ 2020 National Income Classification. The empirical study aims to demonstrate two main issues: (1) whether there is a relationship between the firm’s obstacles and the firm’s bribe payments; and (2) does bribery have an impact on the firm’s operations, in particular on exports? How these impacts will change under the intervention of the levels of obstacles that a firm faces. Estimations are conducted on four subgroups, which the World Bank classifies using the national income. Then, we estimate the models on two subgroups by firm size in the robustness test. The results find a positive relationship of all three barriers, including tax administration, business licensing, and political instability, to bribery payments across all groups classified by national income and firm size. Institutional inefficiencies are the driving force behind corporate bribery. Gaps in administrative management push firms to be willing to engage in bribes. From then they might exchange for advantages. However, these illegal activities of firms are not a solution with long-term benefits. Its consequences not only cause losses to business operations such as reduced labor productivity Dutta and Sobel (2016) but also cause an unhealthy business environment, inhibiting the development of the economy Gründler and Potrafke (2019). Therefore, policymakers can focus on reforming the bureaucracy, shortening cumbersome procedures, increasing support for businesses in administrative procedures such as tax declaration, applying for permits, etc. to combat under-table activities. In particular, for less developed countries, the goal of increasing the effectiveness of the tax administration system should be prioritized. Because the decrease in efficiency of this activity is very meaningful in creating pressure for enterprises to participate in bribery. In addition, the bribery practices of SMEs are more sensitive to institutional inefficiencies. This result can be explained by the well-known characteristics of SMEs such as lack of experience, lack of knowledge, and ability to negotiate with harassment by authorities. Therefore, policies to support SMEs need to be concretized and implemented. In the second analysis of the paper, we find empirical results that support the “greasing the wheel” view of the group of developing countries. This conclusion implies that firms can generate export advantages through lobbying bribes. In contrast, the bribery behavior of firms in developed countries (HI group) does not bring advantages to firms’ exports. The relationship between bribery and exports is negative. In addition, our study found no relationship between bribery and exports in the least developed countries (LDCs). In addition, three problems including competition, lack of skilled human resources, financial constraints are known to be three common hindrances of enterprises in the development process. We find evidence that the effectiveness of informal payments is significantly reduced when firms face such constraints, especially those in middle-income countries and SMEs. However, these results imply that bribery creates unfairness in international market participation among firms. This is a potential risk for the sustainable development of the economy. Therefore, measures to limit corruption, as well as bribery, should be given due attention. These results suggest several policy implications. In which, to ensure equality in Economies 2022,10, 28 23 of 26 the business environment, businesses and the state need to have close coordination. For example, the government needs to overcome cumbersome procedures and legal barriers to create more favorable conditions for businesses to access capital. In addition, human resource training policies to meet the needs of high-skilled workers for export businesses should also be focused on. The strengthened internal strength of the business is also likely to increase the resistance of the business to harassment and illegal solicitation from bureaucrats. Furthermore, internal barriers are more severe in SMEs Bartlett and Bukviˇc (2001). Our paper finds significant effects of bribery payment on exports in this group of firms. Therefore, governments need to have specific policies and pay more attention to this group of businesses. Author Contributions: Conceptualization, T.H.P.; methodology, T.H.P.; software, T.H.P.; formal analysis, T.H.P.; data curation, T.H.P.; writing—review and editing, T.H.P. and R.S.; supervision, R.S. All authors have read and agreed to the published version of the manuscript. Funding: We acknowledge support by the Deutsche Forschungsgemeinschaft (DFG—German Research Foundation) and the Open Access Publishing Fund of Technical University of Darmstadt. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: All data generated or analyzed to support the findings of the present study are included this article. The raw data can be obtained from the authors, upon reasonable request. Conflicts of Interest: The authors declare no conflict of interest. Appendix A Table A1. The list of countries *. Low Income Countries Low-Medium Income Countries Upper-Medium Income Countries Hight Income Countries Afghanistan Angola Albania Barbados Burkina Faso Bangladesh Argentina Belgium Burundi Benin Armenia Chile Central African Republic Bhutan Azerbaijan Croatia Chad Bolivia Belarus Cyprus Drc Cambodia Bosnia and Herzegovina Czech Republic Eritrea Cameroon Botswana Estonia Ethiopia Cape Verde Brazil Greece Gambia Congo Bulgaria Hungary Guinea Côte d’Ivoire Bulgaria Israel Liberia Djibouti China Italy Madagascar Egypt Colombia Latvia Malawi El Salvador Costa Rica Lithuania Mali Ghana Dominica Luxembourg Mozambique Honduras Dominican Republic Malta Niger India Ecuador Mauritius Rwanda Kenya Gabon Panama Sierra Leone Kyrgyzstan Georgia Poland South Sudan Lao PDR Guatemala Portugal Sudan Lesotho Guyana Romania Tajikistan Mauritania Indonesia Slovakia Togo Moldova Iraq Slovenia Uganda Mongolia Jamaica Sweden Yemen Morocco Jordan Trinidad and Tobago Myanmar Kazakhstan Uruguay Nepal Lebanon Nicaragua Malaysia Nigeria Mexico Pakistan Montenegro Papua New Guinea Namibia Philippines North Macedonia Economies 2022,10, 28 24 of 26 Table A1. Cont. Low Income Countries Low-Medium Income Countries Upper-Medium Income Countries Hight Income Countries Senegal Paraguay Sri Lanka Peru Tanzania Russia Timor-Leste Serbia Tunisia South Africa Ukraine Suriname Uzbekistan Thailand Vanuatu Turkey Vietnam Venezuela Zambia Zimbabwe Note: * Country classification by income based on World bank’s publication 2020. Notes 1 Thanks to Thanh et al. (2021) for the idea considering the effect of obstacles. However, they focused on four types of institutional obstacles. In this study, the obstacles come from the firm’s side (the internal obstacles) and are observed more detail in sub-datas. 2 According to Abor (2008), companies with more than 10% foreign equity often have better access to information related to foreign markets. Ownership of more than 10% in companies that demonstrate the right to express an opinion at the general meeting of shareholders can influence decisions in the company’s operations. References Abor, Joshua. 2008. Determinants of the Capital Structure of Ghanaian Firms. Nairobi: African Economic Research Consortium. Asiedu, Elizabeth, and James Freeman. 2009. The effect of corruption on investment growth: Evidence from firms in Latin America, Sub-Saharan Africa, and transition countries. Review of Development Economics 13: 200–14. [CrossRef] Aw, Bee Yan, Sukkyun Chung, and Mark J. Roberts. 2000. Productivity and turnover in the export market: Micro-level evidence from the Republic of Korea and Taiwan (China). The World Bank Economic Review 14: 65–90. [CrossRef] Ayyagari, Meghana, Asli Demirgüç-Kunt, and Vojislav Maksimovic. 2014. Bribe payments and innovation in developing countries: Are innovating firms disproportionately affected? Journal of Financial and Quantitative Analysis 49: 51–75. [CrossRef] Bartlett, Will, and Vladimir Bukviˇc. 2001. Barriers to SME growth in Slovenia. MOST: Economic Policy in Transitional Economies 11: 177–95. [CrossRef] Beck, Thorsten, and Asli Demirguc-Kunt. 2006. Small and medium-size enterprises: Access to finance as a growth constraint. Journal of Banking & Finance 30: 2931–43. Beck, Paul J., and Michael W. Maher. 1989. Competition, regulation and bribery. Managerial and Decision Economics 10: 1–12. [CrossRef] Bellone, Flora, Patrick Musso, Lionel Nesta, and Stefano Schiavo. 2010. Financial constraints and firm export behaviour. World Economy 33: 347–73. [CrossRef] Bernard, Andrew B., J. Bradford Jensen, and Robert Z. Lawrence. 1995. Exporters, jobs, and wages in US manufacturing: 1976–1987. Brookings papers on economic activity. Microeconomics 1995: 67–119. Birhanu, Addis G., Alfonso Gambardella, and Giovanni Valentini. 2016. Bribery and investment: Firm-level evidence from Africa and Latin America. Strategic Management Journal 37: 1865–77. [CrossRef] Box, Marcus. 2008. The death of firms: Exploring the effects of environment and birth cohort on firm survival in Sweden. Small Business Economics 31: 379–93. [CrossRef] Campos, Nauro F., Saul Estrin, and Eugenio Proto. 2010. Corruption As A Barrier to Entry: Theory and Evidence. Available online: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1711074 (accessed on 22 November 2010). Cariolle, Joël, and Petros Sekeris. 2021. How Export Shocks Corrupt: Theory and Evidence. Available online: https://hal.archivesouvertes.fr/hal-03164648/document (accessed on 10 March 2021). Coase, Ronald Harry. 1937. The nature of the firm. Economica 4: 386–405. [CrossRef] Cobham, Alan. 1954. Priority assignment in waiting line problems. Journal of the Operations Research Society of America 2: 70–76. [CrossRef] Cuervo-Cazurra, Alvaro. 2006. Who cares about corruption? Journal of International Business Studies 37: 807–22. [CrossRef] Dal Bó, Ernesto, and Martín A. Rossi. 2007. Corruption and inefficiency: Theory and evidence from electric utilities. Journal of Public Economics 91: 939–62. [CrossRef] De Rosa, Donato, Nishaal Gooroochurn, and Holger Görg. 2010. Corruption and Productivity Firm-Level Evidence from the BEEPS Survey. No. 5348, Policy Research Working Paper. Washington, DC: The World Bank. Dreher, Axel, and Martin Gassebner. 2013. Greasing the wheels? The impact of regulations and corruption on firm entry. Public Choice 155: 413–32. [CrossRef] Dutta, Nabamita, and Russell Sobel. 2016. Does corruption ever help entrepreneurship? Small Business Economics 47: 179–99. [CrossRef]