Effects of bureaucratic corruption on firms' financial constraints

Ezeibekwe, Obinna Franklin Article Effects of bureaucratic corruption on firms' financial constraints The Journal of Entrepreneurial Finance (JEF) Provided in Cooperation with: The Academy of Entrepreneurial Finance (AEF), Los Angeles, CA, USA Suggested Citation: Ezeibekwe, Obinna Franklin (2025) : Effects of bureaucratic corruption on firms' financial constraints, The Journal of Entrepreneurial Finance (JEF), ISSN 2373-1761, Pepperdine University, Graziadio School of Business and Management and The Academy of Entrepreneurial Finance (AEF), Malibu, CA and Los Angeles, CA, Vol. 27, Iss. 1, pp. 1-30, https://doi.org/10.57229/2373-1761.1505 This Version is available at: https://hdl.handle.net/10419/319785 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-nc-nd/4.0/ The Journal of Entrepreneurial Finance The Journal of Entrepreneurial Finance Volume 27 Issue 1 2025 Article 1 4-2025 Effects of Bureaucratic Corruption on Firms' Financial Constraints Effects of Bureaucratic Corruption on Firms' Financial Constraints Obinna Franklin Ezeibekwe Northern Illinois University Follow this and additional works at: https://digitalcommons.pepperdine.edu/jef Part of the Econometrics Commons, Finance Commons, and the Growth and Development Commons Recommended Citation Recommended Citation Ezeibekwe, Obinna Franklin (2025) "Effects of Bureaucratic Corruption on Firms' Financial Constraints," The Journal of Entrepreneurial Finance : Vol. 27: Iss. 1, pp. 1-30. DOI: https://doi.org/10.57229/2373-1761.1505 Available at: https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 This Article is brought to you for free and open access by the Graziadio School of Business and Management at Pepperdine Digital Commons. It has been accepted for inclusion in The Journal of Entrepreneurial Finance by an authorized editor of Pepperdine Digital Commons. For more information, please contact bailey[email protected]. Effects of Bureaucratic Corruption on Firms’ Financial Constraints Obinna Franklin Ezeibekwe Northern Illinois University, DeKalb, IL USA [email protected] Abstract: This study provides the first empirical assessment of the causal impact of bureaucratic corruption on firms’ financial constraints in Nigeria by calculating treatment effects using linear and non-linear estimators to account for potential heterogeneous treatment effects across firm groups. Formally, the theoretical framework models how corruption may facilitate or restrain firms’ financial access by shaping their cost functions, which consequently influences their success or failure and ability to raise the collateral for borrowing. My analysis, using the bivariate probit method and two binary instruments, reveals that corruption significantly increases the probability of a representative MSME and firm being financially constrained by approximately 62 to 64 and 61 to 63 percentage points, respectively. When the IV estimator is utilized to calculate local effects, I find that the effect is about 90 to 91 percentage points for a typical MSME facing obstacles with obtaining business licenses and tax administration, respectively. The effect is 92 percentage points for all firms using both instruments. Furthermore, the results show that Nigerian MSMEs are about 17 to 19 percentage points more likely to be financially constrained than large firms and that corruption’s impact on firms’ access to finance does not depend on firm size. Finally, firms that perceive corruption as a “minor” barrier experience the most difficulty obtaining external finance. This study highlights the severe constraint that corruption poses to Nigerian firms’ access to finance and advocates for regulatory amendments to address issues with the tax administration and the ease of obtaining business licenses and permits. Keywords: Corruption, Financial constraints, Nigeria, MSME, Bivariate probit 1 Ezeibekwe: Bureaucratic Corruption and Firms' Financial ConstraintsPublished by Pepperdine Digital Commons, 2025 1. Introduction Micro, small, and medium enterprises (MSMEs) are crucial to less-developed countries (LDCs) by creating employment opportunities, driving economic growth, and reducing poverty (Maksimov et al., 2017). However, MSMEs require access to sustainable finance to perform these critical developmental functions. Unfortunately, according to a 2021 survey report by the Small and Medium Enterprises Development Agency of Nigeria (SMEDAN) and the National Bureau of Statistics (NBS), access to finance remains the biggest challenge for MSMEs’ growth in Nigeria.1 Firms face financial constraints due to several factors, such as domestic ownership (Blalock et al., 2008), underdeveloped financial architecture (Love, 2003), and a weak legal and institutional environment (Qian and Strahan, 2007). One significant factor that undermines the institutional environment is corruption; however, there are still significant research gaps regarding how corruption impacts MSMEs’ access to finance in LDCs. Therefore, the primary objective of this research is to estimate the treatment effects of corruption on Nigerian firms’ financial constraints by comparing treatment effects estimates derived using linear and non-linear estimators. Corruption, defined as the misuse of entrusted power for private gain, remains a substantial economic and social issue in most LDCs and its systemic nature makes it challenging to address. Economic research provides conflicting predictions regarding the impact of corruption on enterprises’ access to finance. On the one hand, corruption could “sand the wheels” of access to finance in several ways. First, bribes reduce firm growth even more than taxation, making it more difficult for firms to raise collateral and acquire loans from financial institutions (Fisman and Svensson, 2007). Second, corruption depresses economic growth, which could consequently affect the ability of banks to lend (Johnson et al., 2011; Gr u ¨ ndler and Potrafke, 2019; Mo, 2001; M ´ eon and Sekkat, 2005). Third, corruption negatively affects trust (Morris and Klesner, 2010; Banerjee, 2016), which could result in firms doubting their ability to acquire external finance or banks doubting their ability to recover loans in the event of default, thereby preventing them from lending. Fourth, corrupt public officials and agencies could misappropriate funds intended for the benefit of firms or industries or other purposes (Svensson, 2005; Mauro, 1998). On the other hand, corruption could “grease the wheels” of access to finance for firms by reducing bureaucracy and red tape associated with obtaining external finance (Dreher and Gassebner, 2013), reducing the time spent on queues (Lui, 1985), assisting firms to bypass government regulations (Jiang and Nie, 2014), and improving efficiency in countries with weak institutions (M ´ eon and Weill, 2010). Firms operating in Nigeria face several challenges, including multiple taxation, levies, and extortion by public officials and loosely organized gangs known as “area boys.” These challenges, referred to as “lower-level corruption,” result in a significant increase in the cost of doing business. However, “Upper-level corruption” entails the illegal diversion of government resources intended to support firms, thereby undermining business growth and impact. The history of anti-corruption efforts, to address both upperand lower-level corruption, in Nigeria may be traced to the Second Republic when the pioneer anti-corruption agency, the Code of Conduct Bureau (CCB) was established in 1979. The Fourth Republic which started in 1999 witnessed the establishment of the Independent Corrupt Practices Commission (ICPC) in 2000 as a spe1https://smedan.gov.ng/wp-content/uploads/2022/03/2021-MSME-Survey-Report_1.pdf 2The Journal of Entrepreneurial Finance, Vol. 27, Iss. 1 [2025], Art. 1 https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 DOI: 10.57229/2373-1761.1505 cialized agency to tackle public sector corruption, including graft, bribery, and abuse of office. The Economic and Financial Crimes Commission (EFCC) was introduced three years later as a law enforcement agency to investigate financial crimes like money laundering, advanced-fee fraud (known in Nigeria as “419”), and related offenses. These agencies have been instrumental in curbing corruption in Nigeria, but more work needs to be done to create a more enabling environment for businesses to thrive and contribute to Nigeria’s economic growth and development. Additionally, the government has made efforts to ensure that small businesses thrive. For example, SMEDAN was established in 2003 to coordinate the development of the MSME subsector. However, its lack of transparency makes it difficult to study its budgetary practices or identify the beneficiaries of its programs. The Central Bank of Nigeria (CBN) has also launched various programs aimed at supporting MSMEs, such as lending to government development finance institutions to enable MSMEs to access finance at more affordable rates. Notably, the CBN’s Small and Medium Enterprises Equity Investment Scheme (SMEEIS) mandates all banks to reserve 10 percent of their after-tax profit for equity investment or to offer single-digit interest rate loans to small businesses. Similarly, the CBN approved ₦500 billion Naira debenture stock, to be issued to small firms by the Bank of Industry (BOI), and the ₦200 billion Naira Small and Medium Enterprises Credit Guarantee Scheme (SMECGS) was aimed at enhancing credit to MSMEs.2 Nonetheless, the effectiveness of these CBN interventions is reliant on the degree of monitoring carried out by the CBN. These efforts targeted at MSMEs and other firms are aimed at reducing high poverty and unemployment rates. Despite these efforts, poverty and unemployment rates in the country of 211 million people continue to skyrocket. The NBS, in collaboration with the National Social Safety-Nets Coordinating Office (NASSCO), the United Nations Development Programme (UNDP), the United Nations Children’s Fund (UNICEF), and the Oxford Poverty and Human Development Initiative (OPHI), developed the National Multidimensional Poverty Index (NMPI)3 to provide a more comprehensive measure of poverty that considers monetary poverty, education, and basic infrastructure. According to the NMPI, about 63 percent of Nigerians were multidimensionally poor in 2022, with 65 percent of this population living in Northern Nigeria and 72 percent living in rural areas. Additionally, the unemployment rate in Nigeria increased from 14 percent in 2017 to 33 percent in 2020, as reported by the NBS. Because corruption may impact the growth of MSMEs and all firms, this research provides the first empirical assessment of the effects of bureaucratic corruption on Nigerian firms’ financial constraints. Additionally, I estimate the models using data for MSMEs and all firms. Using two binary instruments that are relevant and valid, I employ the instrumental variable estimator to estimate the linear model and a bivariate probit estimator to estimate the average treatment effect (ATE) for compliers and non-compliers. This approach is important because the local effect is unlikely to be equivalent to the ATE when there are heterogeneous treatment effects across firm groups. Additionally, applying different estimators highlights the behavior of different estimators when certain assumptions are made. There are many reasons why this research is essential. First, Nigeria, being the most populous country in Sub-Saharan Africa, is representative of the region and faces similar development challenges such as corruption and an underdeveloped MSME sub-sector, as other countries in the 2 https://www.cbn.gov.ng/Devfin/smefinance.asp 3 https://mppn.org/wp-content/uploads/2022/11/MPI_web_Nov15_FINAL.pdf 3 Ezeibekwe: Bureaucratic Corruption and Firms' Financial ConstraintsPublished by Pepperdine Digital Commons, 2025 region. Therefore, the results of this study could be generalized to other countries in the region. Also, this study is relevant because of the emphasis placed on eradicating corruption and growing MSMEs during Nigeria’s just concluded 2023 general elections. This paper provides policymakers with the first empirical evidence of how firm-related corruption, not corruption in general, affects the economy. Finally, this research would help to determine whether the present and future governments of Nigeria should prioritize anti-corruption campaigns. The second section reviews the literature and the methodology is described in the third section. The fourth section presents the results and the fifth and final section concludes. 2. Literature Review The existing literature examines the effects of corruption on firms’ financial access, but limited research exists for countries in the Sub-Saharan African region. The literature also explores corruption-related factors that may affect firms’ access to finance. The main objective of Xu and Yano (2017) is to estimate the effect of anti-corruption efforts on the financing of and investing in innovation using firm-level data obtained from the China Stock Market and Accounting Research (CSMAR) database of 1895 establishments from 2009 to 2015. Anti-corruption efforts can improve the financing of and investment in innovation by reducing the expropriation problem, which argues that firms are less likely to undertake risky innovative pursuits if they fear that rents accruable from innovation could be expropriated by corrupt public officials thereby increasing the costs associated with innovation. The study employs a probit, GMM, and dynamic GMM model with three external instruments to account for endogeneity and estimate the financial access equation. The results indicate that stronger anticorruption efforts are associated with improved access to finance, particularly long-term debt, for firms. Utilizing the data for 79 countries, Amin and Motta (2023) investigate how bureaucratic corruption affects the access to finance of manufacturing small and medium enterprises (SME). The findings suggest that corruption increases the probability of these firms being financially constrained. The probability of being financially constrained increases by about 4 to 5 percentage points for a standard deviation rise in corruption. The rise in the probability of being financially constrained is smaller in countries where credit bureaus operate and credit markets function more freely. Park (2012) analyzes the impact of corruption on both the banking sector and economic growth, relying on data from 76 countries. The results indicate that corruption exacerbates issues related to non-performing loans in the banking sector. Additionally, corruption hampers economic growth by disrupting the allocation of bank resources, diverting them from viable projects to detrimental ones. This reduces the quality of private investments, consequently leading to a decline in economic growth. Wellalage et al. (2019) study how corruption affects the access to credit for small and medium enterprises (SMEs) in South Asia. The methods used are the probit estimator and the instrumental variable probit to account for endogeneity. Accordingly, the binary dependent variable measures whether an SME is credit-constrained or not. The data is obtained from the 2014 WBES for Afghanistan, Bangladesh, India, Nepal, and Pakistan. The result shows that corruption increases the likelihood of SMEs being credit-constrained by about 7.6 percent. On the contrary, some research provides evidence that bribery may facilitate firms’ access to bank credit facilities. In a study of 14 transition economies, F ung ´ a ˇ cov ´ a et al. (2015) employ firm-level accounting data 4The Journal of Entrepreneurial Finance, Vol. 27, Iss. 1 [2025], Art. 1 https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 DOI: 10.57229/2373-1761.1505 to evaluate how bribery affects firms’ total bank debt ratios, the total amount of bank debt divided by the total amount of assets. Their findings indicate that higher levels of bribery increase the total bank debt ratio, suggesting that bribery stimulates bank lending. Furthermore, bribery facilitates shorter-term bank debt ratios but impedes longer-term bank debt ratios. Another strand of literature studied how factors related to corruption cause firms’ financial constraints. Deploying a sample of 43 countries, Qian and Strahan (2007) regress the terms of loan contracts on some institutional and legal variables. Loan availability increases with stronger creditor rights because banks are more willing to extend cheaper loans with longer maturities if they can be protected during bankruptcy. Also, foreign banks tend to have a higher lending volume in countries with stronger creditor rights. Another study by Love (2003) utilizes data from about 5,000 firms from 36 countries from 1988 to 1998 to investigate how financial development reduces financial constraints. This evidence suggests that financial market development reduces the financial constraints faced by firms, leading to more efficient allocation of resources and enhancing economic growth. This research differs from past studies by being the first to study the treatment effects across groups by comparing treatment effects estimates derived from linear and non-linear methods. In contrast to the study by Xu and Yano (2017), which focused on China, this research paper focuses on Nigeria, which provides a better setting for examining corruption and its impact on firms’ financial constraints for some reasons. First, Nigeria’s corruption level, according to the Corruption Perception Index (CPI) presented by Transparency International, 4 is higher than that of China, despite the anti-corruption campaigns and framework established by different past governments. The CPI ranks countries worldwide on a scale of 0 to 100 based on the perceived level of public sector corruption, where 100 means least perceived public sector corruption and 0 means most perceived public sector corruption. Nigeria’s 2015 and 2022 CPI scores of 26 and 24 are in the 18th and 14th percentile while China’s 2015 and 2022 CPI scores of 37 and 45 lie in the 50th and 64th percentile for the 168 and 180 countries studied, respectively. This evidence shows that corruption has gotten worse in Nigeria over time. Second, Nigeria’s non-bank and non-farm private MSMEs are less developed than those of China. For instance, small firms in China contribute over 60 percent of the country’s gross domestic product, whereas in Nigeria, it is about 48 percent. Moreover, unlike Xu and Yano (2017) that investigated the impact of anti-corruption on the financing of all Chinese firms, this research focuses on how corruption affects the financing of MSMEs, which are more likely to face financial constraints (Oliveira and Fortunato, 2006; Hyytinen and V ¨ a ¨ an ¨ anen , 2006), as well as all firms. In developing the theoretical underpinnings of this study, I build upon the profit maximization problem using a Cobb-Douglas function while demonstrating how corruption increases or decreases the costs of inputs thereby affecting firm growth and consequently impacting firms’ ability to access external finance. 2.1. Theoretical Framework This study employs a profit maximization model to examine how corruption affects firms’ financial constraints. Since corruption can either sand or grease the wheels of financial access, this study’s theoretical framework formally demonstrates how corruption could hinder or stimulate 4 https://www.transparency.org/en/cpi/2022 5 Ezeibekwe: Bureaucratic Corruption and Firms' Financial ConstraintsPublished by Pepperdine Digital Commons, 2025 ∂C firms’ access to finance by affecting firm growth. When firms have to pay bribes to operate normally, access government business support services, or secure contracts, their costs of inputs and business expenses increase, which undermines their growth and ability to obtain the collateral for accessing finance. However, in some cases, corruption can decrease input costs for firms, allowing them to circumvent compliance costs, get preferential treatment, receive more subsidies, or access resources at cheaper rates. These advantages can enable firms to reduce costs and allocate resources more efficiently. I employ a Cobb-Douglas production function that accounts for two inputs, namely capital (K) and labor (L). This function approximates an actual production function and is easy to analyze. The production function of a representative profit-maximizing Nigerian firm, with Y ∈ R ≥ 0, K ∈ R ≥ 0, and L ∈ R ≥ 0, is given as follows. Y = K α L β , 0 < α < 1 and 0 < β < 1 (1) I assume that output increases by less than the proportional change as all inputs. Also, firms have no initial funds and must acquire all the necessary inputs through borrowing. The total cost (TC) function is: TC = wL + rK (2) The costs of inputs L and K are represented by w and r, respectively. I represent the effect of corruption as an additional cost or benefit (E) incurred due to corruption and it is proportional to the level of corruption (C) in the economy. E = ϕC, ϕ ∈ R and C ∈ R>0 (3) The coefficient ϕ denotes the magnitude of the effect of corruption on input costs. The conditions, C ∈ R>0 and r > | ϕC | < w, ensure that the corruption level is positive and input costs are never zero or negative. Computing ∂E gives ϕ. When ϕ > 0, corruption leads to an increase in input costs. Firms reveal their higher level of financial constraints due to corruption by choosing suboptimal input levels. Conversely, when ϕ < 0, corruption results in a decrease in input costs. This causes firms to reveal their lower level of financial constraints by choosing more optimal input levels. When ϕ = 0, firms face financial constraints that are unrelated to corruption. Adjusting the cost function to account for the influence of corruption gives: TC = (w + ϕC)L + (r + ϕC)K (4) I assume that ϕC is identical for all inputs. The optimization problem, subject to the input cost being less than or equal to the budget constraint, is: max K, L pK α L β − (r + ϕC)K − (w + ϕC)L (5) Taking the first partial derivative of (5) with respect to K and L yields: ∂π = αp 𝐾 α−1 L β = r + ϕC (6) ∂K ∂π = βp 𝐾 α 𝐿 β−1 = w + ϕC (7) ∂L Dividing (6) by (7) and solving for L, the value of L can be obtained. β(r + ϕC) L = α(w + ϕC) K (8) 6The Journal of Entrepreneurial Finance, Vol. 27, Iss. 1 [2025], Art. 1 https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 DOI: 10.57229/2373-1761.1505 ∂C ∂C Plugging (8) into (6) produces the optimal capital input, K ∗ . 𝐾∗= ( 𝑝α 𝑟+ϕ𝐶)1−β 1−α−β (𝑝β 𝑤+ϕ𝐶)β 1−α−β (9) Plugging (9) into (8) gives the optimal labor input, 𝐿∗= ( 𝑝α 𝑟+ϕ𝐶)α 1−α−β (𝑝β 𝑤+ϕ𝐶)1−α 1−α−β (10) All things equal, the optimal input levels rise if ϕ < 0, fall if ϕ > 0, and are unaffected by corruption if ϕ = 0. Plugging the optimal inputs into (1) gives the optimal output, Y ∗ , in the presence of corruption. Expressing financial constraint F, which is a variable greater than or equal to zero and zero if negative, as the difference between profits in the absence of corruption and profits in the presence of corruption is as follows: F = Π( Y ˆ ) − Π(Y ∗ , C) (11) F is positive when Π( Y ˆ ) > Π(Y ∗ , C) and zero when Π( Y ˆ ) < Π(Y ∗ , C) or Π( Y ˆ ) = Π(Y ∗ , C), suggesting that there is no positive financial constraint induced by corruption. Taking the partial derivative of (11) to understand the impact of corruption on financial constraint yields: 𝜕𝐹 𝜕𝐶 = 𝜙× 𝑝((𝛼+𝛽)𝜙𝐶+𝛼𝑤+𝛽𝑟) (𝑟+𝜙𝐶)(𝑤+𝜙𝐶)(𝑟+𝜙𝐶 𝑝𝛼 )𝛼 1−𝛼−𝛽(𝑤+𝜙𝐶 𝑝𝛽 )𝛽 1−𝛼−𝛽 ⏟ + (12) The model predicts that financial constraints become more stringent when the effect of corruption is positive. This means that the partial derivative of financial constraint (F) with respect to corruption (C) is positive when ϕ is greater than zero. Corruption increases the cost of inputs for firms, which hinders their profitability and growth. Consequently, it becomes more difficult for them to raise the required collateral to access external finance. On the other hand, financial constraints weaken when the effect of corruption is negative, ϕ < 0. This implies that ∂F is negative. Corrupt practices reduce input costs by allowing them to bypass compliance costs or receive preferential treatments, which encourages firm growth. This makes it easier for firms to acquire the collateral necessary to secure external financing. When corruption has no effect (ϕ = 0), ∂F becomes zero, indicating that financial constraints are not affected by corruption. This leads to the hypothesis that higher levels of corruption increase financial constraints for Nigerian firms, which I tested in my empirical research. The theoretical framework provides the economic interpretation of the estimates of corruption provided by the empirical analysis. 3. Methodology 3.1. Data To investigate the causal impact of corruption on firms’ access to finance in Nigeria, I analyze survey data from the 2014-2015 World Bank Enterprise Survey (WBES) conducted in Nigeria between April 2014 and February 2015. The WBES is a comprehensive and nationally representative survey of non-agricultural and non-financial private enterprises. The sample of 7 Ezeibekwe: Bureaucratic Corruption and Firms' Financial ConstraintsPublished by Pepperdine Digital Commons, 2025 2003). For example, the absence of a transparent and efficient tax administration and business licensing creates opportunities for corrupt practices. Firms experiencing these impediments are more likely to seek assistance from government agencies, exposing them to corrupt officials who may demand bribes. Therefore, the instruments can predict the treatment status, Corruption, once the other exogenous covariates have been partialled out. This condition can be assessed empirically using the weak instrument test. In addition, the instruments have no direct relationship with firms’ financial access except through the first stage. Both instruments are primarily influenced by institutions, regulations, and administrative procedures, rather than financial constraints. Additionally, banks do not inquire about whether a firm is facing obstacles with the tax administration (this does not mean that the firm does not pay taxes) and obtaining licenses before deciding whether to approve or not approve a loan. The instruments are not correlated with any omitted factor in ϵi and vi, such as the firm’s profitability. The Sargan test, with a null hypothesis that the system has a solution, tests that all restrictions are internally consistent. 4. Results and Discussion 4.1. Main Findings The linear models are estimated using the least squares method, while the non-linear models are estimated using maximum likelihood. The IV is estimated using one instrument and the 2SLS is estimated using two instruments. The IV estimator calculates the average treatment effect for compliers using a binary instrument (Imbens and Angrist, 1994; Angrist et al., 1996; Angrist and Pischke, 2009). This examination utilizes the ordinary least square (OLS), IV estimator, and 2SLS estimators to estimate the linear model and adopts the probit and bivariate probit to estimate the non-linear models. The standard errors for the OLS, IV, and 2SLS are heteroskedasticity-consistent (HC3), as indicated in Long and Ervin (2000). For the bivariate probit estimator, the standard errors were calculated by bootstrapping confidence intervals using 10,000 simulated coefficient vectors from the posterior distribution of the estimated model parameters. The estimates for ρ are significant in all cases. The variance inflation factor (VIF) was used to assess multicollinearity, and the result is reported in Table A.1. All the values are close to 1, the smallest possible value for VIF, indicating that multicollinearity is not a problem. The first-stage result, reported in Table 5, indicates that both Tax Admin and License have positive, significant, and robust effects on Corruption. This holds whether they are used separately and together in regression and after adjusting for heteroskedasticity. The signs of the instruments from the first-stage regressions are positive and align with the hypothesized sign. The first-stage positive effects of Tax Admin and License on the probability of facing corruption are about 39 and 22 percentage points, respectively. The F-statistics from the three regressions are significant at the 1 percent level. The F-statistic increases from 3.7 by about 527 percent, 300 percent, and 559 percent when Tax Admin, License, and Tax Admin and License combined are included in the regressions, respectively. 14The Journal of Entrepreneurial Finance, Vol. 27, Iss. 1 [2025], Art. 1 https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 DOI: 10.57229/2373-1761.1505 Table 5: First-Stage Results Dependent variable: Corruption Ordinary Least Square (OLS) (1) (2) (3) (4) Tax Admin License 0.338 ∗∗∗ (0.031) 0.218 ∗∗∗ 0.275 ∗∗∗ (0.033) 0.116 ∗∗∗ Capital City − 0.029 − 0.009 (0.023) − 0.025 (0.023) − 0.010 Business City South (0.019) 0.094 ∗∗∗ (0.017) − 0.047 ∗∗ (0.020) (0.018) 0.074 ∗∗∗ (0.016) − 0.018 (0.019) (0.018) 0.075 ∗∗∗ (0.016) − 0.038 ∗∗ (0.019) (0.018) 0.068 ∗∗∗ (0.016) − 0.019 (0.019) Manufacturing − 0.001 0.004 0.001 0.004 (0.020) (0.018) (0.019) (0.018) Retail 0.005 0.009 0.007 0.010 (0.024) (0.022) (0.023) (0.022) Firm Age − 0.0001 − 0.0002 0.001 0.0002 Revenue Innovation Sole Proprietor (0.001) − 0.004 (0.004) − 0.0003 (0.017) − 0.009 (0.020) (0.001) − 0.002 (0.004) − 0.022 (0.015) − 0.006 (0.020) (0.001) − 0.004 (0.004) − 0.004 (0.016) − 0.002 (0.020) (0.001) − 0.002 (0.004) − 0.020 (0.015) − 0.003 (0.020) Manager Education 0.006 ∗∗ 0.005 0.005 ∗ 0.004 (0.003) (0.003) (0.003) (0.003) Female Owner 0.009 0.005 − 0.003 − 0.001 (0.022) (0.021) (0.021) (0.021) Subsidiary Firm 0.038 ∗ 0.028 0.033 ∗ 0.027 Constant (0.020) 0.799 ∗∗∗ (0.080) (0.019) 0.518 ∗∗∗ (0.079) (0.020) 0.650 ∗∗∗ (0.080) (0.019) 0.492 ∗∗∗ (0.079) Observations 1,696 1,696 1,696 1,696 R2 0.026 0.152 0.103 0.169 Adjusted R2 0.019 0.145 0.096 0.162 Residual Std. Error 0.340 0.317 0.326 0.314 F Statistic 3.706 ∗∗∗ 23.169 ∗∗∗ 14.802 ∗∗∗ 24.449 ∗∗∗ Note: ∗ p < 0.1; ∗∗ p < 0.05; ∗∗∗ p < 0.01 The numbers in round brackets () are standard errors. Standard errors are heteroskedasticity-consistent (HC3) 15 Ezeibekwe: Bureaucratic Corruption and Firms' Financial ConstraintsPublished by Pepperdine Digital Commons, 2025 4.1.1. Effects of Bureaucratic Corruption on MSMEs’ Financial Constraints The effects of bureaucratic corruption on MSMEs’ financial constraints using OLS, IV, 2SLS, probit, and bivariate probit methods are reported here. According to the OLS estimator that assumes that Corruption is exogenous, MSMEs (used interchangeably with small firms) that identified corruption as an obstacle are, on average, approximately 0.318 more likely to be financially constrained than those that did not identify corruption as an obstacle. Bribery increases input costs and inhibits firm growth, thereby making it more difficult for firms to obtain the collateral required for accessing external finance. The estimate of the non-linear estimator that assumes exogeneity is only 0.3 percent larger but they are the same to two decimal places. The probit estimator is also more efficient. According to Table 6, the effect of Corruption nearly triples when either or both instruments are used. Although the IV and 2SLS estimates are similar, the IV using Tax Admin as an instrument has the largest estimate. The 2SLS estimate is closer to the IV estimate using Tax Admin because the IV estimator has a stronger first stage. Utilizing Tax Admin as an instrument, corruption is estimated to raise the likelihood of facing financial constraints for an MSME experiencing tax administration hindrances by 91 percentage points. Using the same instrument, the bivariate probit suggests that corruption increases the probability of a representative MSME being financially constrained by roughly 62 percentage points. The estimate derived from the linear IV estimator is approximately 47 percent higher than the non-linear estimate. For models using License as an instrument, if an MSME is grappling with obtaining business licenses and permits, it is about 90 percentage points more likely to be financially constrained due to corruption. The corresponding bivariate probit estimate indicates a 64 percentage point rise in the probability of becoming constrained financially. The IV estimate is 40 percent higher than the bivariate probit estimate. The theoretical framework provides the economic interpretation of the effect. The positive estimates imply that corruption increases firm costs, as a result, reducing their growth and ability to raise collateral for borrowing. As expected, IVs and 2SLS estimators are less efficient than the OLS, but the 2SLS is more efficient than the IV estimators. The bivariate probit estimators are more efficient than their linear counterparts, whether one or two instruments are utilized. Also, the probit method has a superior explanatory power than the OLS. The Wu-Hausman tests suggest that Corruption is indeed endogenous, and the weak instrument tests confirm the relevance of the instruments used in all models. The 2SLS is used to compute the Sargan test, which suggests that the system has a solution. The statistically significant ρ ˆ estimates suggest the existence of omitted variable bias in the system of equations. Therefore, the OLS and probit estimators are inconsistent. In Table 6, six covariates produce statistically significant estimates across the linear and non-linear methods: Capital City, Business City, South, Retail, Firm Age, and Manager Education. The analysis shows that there is geographical variation in financial constraints, with MSMEs located in the capital city experiencing higher levels of financial constraints. This could be due to their proximity to public officials, which increases the likelihood of paying bribes and becoming bankrupt. Additionally, higher operational costs and competition contribute to this problem. Firms in the capital cities tend to face higher rents, labor costs, and operational expenses, which require more capital to cover. Capital cities attract a large number of businesses, leading to increased competition and higher financial pressure to invest in marketing, research and development, and employee benefits to remain competitive. In addition, intense competition for funding 16The Journal of Entrepreneurial Finance, Vol. 27, Iss. 1 [2025], Art. 1 https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 DOI: 10.57229/2373-1761.1505 Table 6: Effects of Bureaucratic Corruption on MSMEs’ Financial Constraints Dependent variable: Financial Constraints OLS (1) IV (2) IV (3) 2SLS (4) Probit (5) (6) Bivariate Probit (7) (8) Corruption 0.318 ∗∗∗ 0.907 ∗∗∗ 0.896 ∗∗∗ 0.903 ∗∗∗ 0.319 ∗∗∗ 0.617 ∗∗∗ 0.637 ∗∗∗ 0.623 ∗∗∗ (0.035) (0.113) (0.126) (0.102) (0.034) (0.043) (0.050) (0.041) Capital City 0.101 ∗∗∗ 0.118 ∗∗∗ 0.117 ∗∗∗ 0.118 ∗∗∗ 0.099 ∗∗∗ 0.092 ∗∗∗ 0.090 ∗∗∗ 0.092 ∗∗∗ (0.023) (0.026) (0.026) (0.026) (0.022) (0.019) (0.026) (0.019) Business City − 0.077 ∗∗∗ − 0.133 ∗∗∗ − 0.132 ∗∗∗ − 0.132 ∗∗∗ − 0.075 ∗∗∗ − 0.083 ∗∗∗ − 0.083 ∗∗∗ − 0.083 ∗∗∗ (0.022) (0.027) (0.027) (0.026) (0.022) (0.019) (0.019) (0.019) South − 0.079 ∗∗∗ − 0.051 ∗ − 0.052 ∗ − 0.051 ∗ − 0.078 ∗∗∗ − 0.055 ∗∗∗ − 0.053 ∗∗∗ − 0.053 ∗∗∗ (0.024) (0.027) (0.027) (0.027) (0.022) (0.019) (0.020) (0.020) Manufacturing 0.036 0.036 0.036 0.036 0.036 0.031 0.030 0.030 (0.023) (0.027) (0.026) (0.027) (0.022) (0.019) (0.019) (0.019) Retail 0.076 ∗∗∗ 0.074 ∗∗ 0.074 ∗∗ 0.074 ∗∗ 0.075 ∗∗∗ 0.065 ∗∗∗ 0.062 ∗∗∗ 0.064 ∗∗∗ (0.028) (0.031) (0.031) (0.031) (0.025) (0.023) (0.022) (0.023) Firm Age 0.004 ∗∗∗ 0.005 ∗∗∗ 0.005 ∗∗∗ 0.005 ∗∗∗ 0.004 ∗∗∗ 0.004 ∗∗∗ 0.004 ∗∗∗ 0.004 ∗∗∗ (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Revenue 0.005 0.008 0.008 0.008 0.006 0.007 0.007 0.007 (0.005) (0.005) (0.005) (0.005) (0.005) (0.005) (0.005) (0.005) Innovation 0.006 0.006 0.006 0.006 0.006 0.002 0.003 0.002 (0.020) (0.022) (0.022) (0.022) (0.019) (0.017) (0.017) (0.017) Sole Proprietor − 0.025 − 0.019 − 0.019 − 0.019 − 0.020 − 0.015 − 0.014 − 0.014 (0.025) (0.028) (0.028) (0.028) (0.024) (0.021) (0.021) (0.021) Manager Education − 0.005 ∗ − 0.009 ∗∗ − 0.009 ∗∗ − 0.009 ∗∗ − 0.005 − 0.004 ∗∗∗ − 0.004 ∗∗∗ − 0.004 ∗∗∗ (0.003) (0.004) (0.004) (0.004) (0.003) (0.001) (0.001) (0.001) Female Owner 0.032 0.027 0.027 0.027 0.029 0.024 0.023 0.027 (0.026) (0.029) (0.029) (0.029) (0.025) (0.022) (0.021) (0.024) Subsidiary Firm 0.004 − 0.018 − 0.018 − 0.018 0.004 − 0.006 − 0.007 − 0.007 (0.025) (0.027) (0.027) (0.027) (0.024) (0.021) (0.021) (0.021) Constant 0.459 ∗∗∗ − 0.012 − 0.003 − 0.009 (0.095) (0.133) (0.142) (0.127) Observations 1,696 1,696 1,696 1,696 1,696 1,696 1,696 1,696 (Pseudo) R2 0.102 (0.104) Weak IV (p) 2 × 10 − 16 2 × 10 − 16 2 × 10 − 16 Wu-Hausman (p) 5 × 10 − 16 2 × 10 − 9 2 × 10 − 16 Sargan (p) 0.918 ρ ˆ (Std. Err.) -0.6(0.1) -0.6(0.1) -0.6(0.1) Note: ∗ p < 0.1; ∗∗ p < 0.05; ∗∗∗ p < 0.01 The instrument for (2) and (6) is Tax Admin, the instrument for (3) and (7) is License, and the instruments for (4) and (8) are Tax Admin and License. Standard errors for (1), (2), (3), and (4) are heteroskedasticity-consistent (HC3). Standard errors for the bivariate probit were calculated from confidence intervals bootstrapped using 10,000 simulated coefficient vectors from the posterior distribution of the estimated model parameters. (5)-(8) are marginal effects. The numbers in round brackets () are standard errors and p means p-value. All equations were estimated with an intercept. 17 Ezeibekwe: Bureaucratic Corruption and Firms' Financial ConstraintsPublished by Pepperdine Digital Commons, 2025 can make it harder for individual firms to secure the necessary capital, leading to greater financial constraints. MSMEs located in business cities typically have better access to financing. This is because such cities have a larger concentration of financial institutions, such as banks and investment firms, which makes it easier for businesses to access financial products and services. Being located in a business city can also attract more investors and customers, leading to more funding and revenue for firms to meet their financial needs and grow. As expected, MSMEs in southern Nigeria have better access to finance due to their proximity to capital markets and the willingness of investors to invest in more prosperous regions with perceived stability and potential for higher returns. The superior consumer spending in southern Nigeria could enhance the revenue and profitability of these firms, making it easier for them to obtain loans. Financial constraints are more prevalent in the retail sub-sector than in the manufacturing or other services sub-sectors. This is probably due to the intense competition that exists in the retail industry, which results in businesses having lower profits and limited financial power. As a result, they may struggle to cover expenses, leading to lower profitability. Additionally, older firms face challenges in obtaining finance as they are more likely to have accumulated debt over the years due to expansions or acquisitions. Servicing debts may limit their ability to take on additional debt, and the payment of dividends to shareholders may further limit the funds available for expansion. On the other hand, firms with more educated managers have lower financial constraints. This can be attributed to their superior financial literacy, which leads to better financial decision-making. This contributes to the financial stability of the firms they manage and reduces the chances of being financially constrained. Furthermore, more educated managers may also have a broader professional network and better knowledge of financial institutions, making it easier for them to secure external funding when needed. 4.1.2. Effects of Bureaucratic Corruption on Firms’ Financial Constraints Here, I add the 110 large firms with complete data to the 1,696 MSMEs and analyze the data for all firms to understand the overarching patterns across enterprises. I include a binary variable, MSME, which equals one if the establishment identifies as an MSME and zero otherwise. The results for all firms are in Table 7. The results provide the treatment effects using linear and non-linear methods. Based on the IV estimates, corruption increases the probability of being financially constrained for a typical Nigerian firm facing impediments related to tax administration and obtaining business licenses and permits by approximately 92 percentage points. Using Tax Admin and License as instruments, the bivariate probit estimates imply that corruption increases the probability of being financially constrained by 61 to 63 percentage points, respectively. After including large enterprises with 100 or more employees, the estimates from the linear method are approximately two percent larger than the MSME IV estimates, using Tax Admin and License as instruments separately in regressions. In contrast, the bivariate probit estimate declines by one percent for Tax Admin and two percent for License. The economic interpretations of these estimates are based on the predictions of the theoretical framework. Including large enterprises with 100 or more employees did not substantially change the estimates. However, this could be due to the limited representation of these firms in the sample. Another reason could be that the effect of corruption is homogeneous across firm sizes. Furthermore, the diagnostic result remains stable after including large firms. 18The Journal of Entrepreneurial Finance, Vol. 27, Iss. 1 [2025], Art. 1 https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 DOI: 10.57229/2373-1761.1505 Table 7: Effects of Bureaucratic Corruption on Firms’ Financial Constraints Dependent variable: Financial Constraints OLS (1) IV (2) IV (3) 2SLS (4) Probit (5) Bivariate Probit (6) (7) (8) Corruption 0.311 ∗∗∗ 0.921 ∗∗∗ 0.915 ∗∗∗ 0.919 ∗∗∗ 0.312 ∗∗∗ 0.611 ∗∗∗ 0.626 ∗∗∗ 0.616 ∗∗∗ (0.034) (0.111) (0.125) (0.101) (0.034) (0.041) (0.047) (0.038) MSME 0.212 ∗∗∗ 0.193 ∗∗∗ 0.193 ∗∗∗ 0.193 ∗∗∗ 0.211 ∗∗∗ 0.169 ∗∗∗ 0.166 ∗∗∗ 0.166 ∗∗∗ (0.050) (0.055) (0.055) (0.055) (0.052) (0.046) (0.045) (0.045) Capital City 0.102 ∗∗∗ 0.122 ∗∗∗ 0.122 ∗∗∗ 0.122 ∗∗∗ 0.100 ∗∗∗ 0.094 ∗∗∗ 0.093 ∗∗∗ 0.094 ∗∗∗ (0.022) (0.025) (0.025) (0.025) (0.021) (0.019) (0.019) (0.019) Business City − 0.073 ∗∗∗ − 0.124 ∗∗∗ − 0.124 ∗∗∗ − 0.124 ∗∗∗ − 0.070 ∗∗∗ − 0.077 ∗∗∗ − 0.077 ∗∗∗ − 0.078 ∗∗∗ (0.022) (0.025) (0.026) (0.025) (0.022) (0.019) (0.019) (0.019) South − 0.085 ∗∗∗ − 0.051 ∗∗ − 0.051 ∗∗ − 0.051 ∗∗ − 0.083 ∗∗∗ − 0.057 ∗∗∗ − 0.056 ∗∗∗ − 0.056 ∗∗∗ (0.023) (0.026) (0.026) (0.026) (0.022) (0.019) (0.019) (0.019) Manufacturing 0.035 0.036 0.036 0.036 0.035 ∗ 0.030 0.030 0.030 (0.023) (0.026) (0.026) (0.026) (0.021) (0.019) (0.019) (0.019) Retail 0.081 ∗∗∗ 0.078 ∗∗ 0.078 ∗∗ 0.078 ∗∗ 0.080 ∗∗∗ 0.069 ∗∗ 0.066 ∗∗ 0.068 ∗∗ (0.028) (0.031) (0.031) (0.031) (0.025) (0.022) (0.022) (0.022) Firm Age 0.005 ∗∗∗ 0.005 ∗∗∗ 0.005 ∗∗∗ 0.005 ∗∗∗ 0.005 ∗∗∗ 0.004 ∗∗∗ 0.005 ∗∗∗ 0.004 ∗∗∗ (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Revenue 0.002 0.004 0.004 0.004 0.002 0.003 0.003 0.003 (0.004) (0.005) (0.005) (0.005) (0.004) (0.004) (0.004) (0.004) Innovation 0.005 0.008 0.008 0.008 0.005 0.002 0.004 0.002 (0.019) (0.022) (0.022) (0.022) (0.019) (0.017) (0.017) (0.016) Sole Proprietor − 0.029 − 0.015 − 0.015 − 0.015 − 0.024 − 0.015 − 0.015 − 0.015 (0.024) (0.027) (0.027) (0.027) (0.023) (0.021) (0.020) (0.020) Manager Education − 0.005 − 0.009 ∗∗ − 0.009 ∗∗ − 0.009 ∗∗ − 0.004 − 0.004 ∗∗∗ − 0.004 ∗∗∗ − 0.004 ∗∗∗ (0.003) (0.004) (0.004) (0.004) (0.003) (0.001) (0.001) (0.001) Female Owner 0.036 0.034 0.034 0.034 0.031 0.027 0.026 0.026 (0.026) (0.029) (0.029) (0.029) (0.024) (0.021) (0.021) (0.021) Subsidiary Firm − 0.005 − 0.028 − 0.028 − 0.028 − 0.005 − 0.014 − 0.014 − 0.014 (0.024) (0.026) (0.027) (0.026) (0.023) (0.020) (0.020) (0.021) Constant 0.292 ∗∗∗ − 0.191 − 0.186 − 0.190 (0.109) (0.143) (0.152) (0.139) Observations 1,806 1,806 1,806 1,806 1,806 1,806 1,806 1,806 R2 (Pseudo R2) 0.113 (0.114) Weak IV (p) 2 × 10 − 16 2 × 10 − 16 2 × 10 − 16 Wu-Hausman (p) 2 × 10 − 16 2 × 10 − 10 2 × 10 − 16 Sargan (p) 0.956 ρ ˆ (Std. Err.) -0.6(0.1) -0.6(0.1) -0.6(0.1) Note: ∗ p < 0.1; ∗∗ p < 0.05; ∗∗∗ p < 0.01 The instrument for (2) and (6) is Tax Admin, the instrument for (3) and (7) is License, and the instruments for (4) and (8) are Tax Admin and License. Standard errors for (1), (2), (3), and (4) are heteroskedasticity-consistent (HC3). The numbers in round brackets () are standard errors. Standard errors for the bivariate probit were calculated from confidence intervals bootstrapped using 10,000 simulated coefficient vectors from the posterior distribution of the estimated model parameters. (5)-(8) are marginal effects. The numbers in round brackets () are standard errors and p means p-value. All equations were estimated with an intercept. 19 Ezeibekwe: Bureaucratic Corruption and Firms' Financial ConstraintsPublished by Pepperdine Digital Commons, 2025 Furthermore, Nigerian MSMEs are about 17 to 19 percentage points more likely to be financially constrained than large firms. One way to explain this effect is that small businesses have lower total revenue than large firms, indicating that smaller businesses are more likely to struggle to raise the collateral needed to obtain bank loans. 4.1.3. Analyzing Firm Size and Industry Heterogeneity and the Degrees of Corruption Having observed the changes in the effect of Corruption after including large firms, I employ an interaction term to assess if the difference in financial constraints between MSMEs and large firms due to corruption is statistically significant. Table 8 shows the result for the interaction term between Corruption and MSME. The effect of the interaction term is not statistically significant, suggesting that the impact of corruption on financial constraints faced by Nigerian firms does not depend on firm size. Similarly, there is no evidence that the impact of corruption differs across industries (Table 9). However, the main conclusion that corruption restricts firms’ access to finance remains unchanged. Additionally, I employ the IV estimator to estimate the effects of the degrees of corruption on firms’ access to finance. I retained Financial Constraints as a binary variable. The ordinal variables Corruption, Tax Admin, and License were divided into five binary variables, and the survey classifications are used to categorize the degree of corruption, tax administration challenges, and issues with obtaining licenses where 0 signifies no obstacle, 1 represents a minor obstacle, 2 indicates a moderate obstacle, 3 suggests a major obstacle, and 4 shows a very severe obstacle. The summary statistics of the new variables are presented in Table A.2. Given the endogeneity of Corruption, I utilize Tax Admin (Minor) and License (Minor) as instruments for Corruption (Minor) and so on. The results of the analysis, as presented in Table 10, indicate that firms that reported corruption as a “minor” barrier experience the highest level of financial constraints for both instruments. One would expect enterprises that perceive corruption as a “very severe” hindrance to their operation to have the largest estimated effect. This result may support the notion stated earlier that while ordinal variables allow the measurement of different levels of intensity, the interpretation of the different categories may be more subjective than that of a binary variable. Overall, these findings underscore the pervasive and deteriorating impact of bureaucratic corruption on Nigerian firms and highlight the need for effective measures to combat corruption and reduce its adverse effects on firms and, consequently, economic growth and development. 20The Journal of Entrepreneurial Finance, Vol. 27, Iss. 1 [2025], Art. 1 https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 DOI: 10.57229/2373-1761.1505 Table 8: Effects of Corruption on Firms’ Financial Constraints By Firm Size Dependent variable: Financial Constraints OLS IV IV (1) (2) (3) Corruption 0.241 ∗ (0.132) 1.146 ∗∗ (0.500) 1.326 ∗ (0.770) MSME 0.147 0.402 0.567 (0.129) (0.456) (0.687) Corruption × MSME 0.075 − 0.243 − 0.436 Capital City Business City South (0.137) 0.101 ∗∗∗ (0.022) − 0.074 ∗∗∗ (0.022) − 0.086 ∗∗∗ (0.023) (0.511) 0.123 ∗∗∗ (0.025) − 0.121 ∗∗∗ (0.026) − 0.049 ∗ (0.027) (0.780) 0.124 ∗∗∗ (0.026) − 0.119 ∗∗∗ (0.027) − 0.048 ∗ (0.028) Manufacturing 0.035 0.036 0.036 Retail Firm Age (0.023) 0.081 ∗∗∗ (0.028) 0.005 ∗∗∗ (0.001) (0.026) 0.078 ∗∗ (0.031) 0.005 ∗∗∗ (0.001) (0.026) 0.077 ∗∗ (0.031) 0.005 ∗∗∗ (0.001) Revenue 0.002 0.004 0.004 (0.004) (0.005) (0.005) Innovation 0.005 0.009 0.010 Sole Proprietor Manager Education (0.019) − 0.030 (0.024) − 0.005 (0.003) (0.022) − 0.012 (0.029) − 0.009 ∗∗ (0.004) (0.023) − 0.009 (0.030) − 0.009 ∗∗ (0.004) Female Owner 0.035 0.035 0.036 Subsidiary Firm Constant (0.025) − 0.004 (0.024) 0.354 ∗∗ (0.157) (0.030) − 0.028 (0.027) − 0.391 (0.464) (0.030) − 0.028 (0.027) − 0.550 (0.691) Observations 1,806 1,806 1,806 R2 0.113 Weak IV (p) 2 × 10 − 16 2 × 10 − 16 Weak IV (p) 2 × 10 − 16 2 × 10 − 16 Wu-Hausman (p) 2 × 10 − 16 1 × 10 − 9 Note: ∗ p<0.1; ∗∗ p<0.05; ∗∗∗ p<0.01 The instrument for (2) is Tax Admin and the instrument for (3) is License. Standard errors are heteroskedasticity-consistent (HC3). The numbers in round brackets () are standard errors and p means p-value. All equations were estimated with an intercept. 21 Ezeibekwe: Bureaucratic Corruption and Firms' Financial ConstraintsPublished by Pepperdine Digital Commons, 2025 Table 9: Effects of Corruption on Firms’ Financial Constraints by Industry Dependent variable: Financial Obstacle LPM IV IV (1) (2) (3) Corruption MSME 0.234 ∗∗∗ (0.062) 0.209 ∗∗∗ (0.050) 0.781 ∗∗∗ (0.211) 0.190 ∗∗∗ (0.055) 0.893 ∗∗∗ (0.234) 0.192 ∗∗∗ (0.056) Corruption × Manufacturing 0.083 0.124 − 0.041 (0.077) (0.260) (0.280) Corruption × Retail 0.199 ∗ 0.374 0.245 Capital City Business City South (0.103) 0.103 ∗∗∗ (0.022) − 0.075 ∗∗∗ (0.022) − 0.084 ∗∗∗ (0.023) (0.295) 0.124 ∗∗∗ (0.025) − 0.126 ∗∗∗ (0.025) − 0.049 ∗ (0.026) (0.365) 0.122 ∗∗∗ (0.025) − 0.124 ∗∗∗ (0.026) − 0.049 ∗ (0.027) Manufacturing − 0.037 − 0.071 0.071 Retail Firm Age (0.074) − 0.092 (0.100) 0.004 ∗∗∗ (0.001) (0.233) − 0.248 (0.268) 0.005 ∗∗∗ (0.001) (0.249) − 0.135 (0.328) 0.005 ∗∗∗ (0.001) Revenue 0.002 0.005 0.005 (0.004) (0.005) (0.005) Innovation 0.007 0.012 0.010 Sole Proprietor Manager Education (0.019) − 0.030 (0.024) − 0.005 (0.003) (0.022) − 0.017 (0.028) − 0.009 ∗∗ (0.004) (0.022) − 0.014 (0.028) − 0.009 ∗∗ (0.004) Female Owner 0.034 0.031 0.032 Subsidiary Firm Constant (0.026) − 0.006 (0.024) 0.366 ∗∗∗ (0.121) (0.029) − 0.030 (0.026) − 0.062 (0.226) (0.029) − 0.029 (0.027) − 0.169 (0.245) Observations 1,806 1,806 1,806 R2 0.116 Weak IV (p) 2 × 10 − 16 2 × 10 − 16 Weak IV (p) 2 × 10 − 16 2 × 10 − 16 Weak IV (p) 2 × 10 − 16 2 × 10 − 16 Wu-Hausman (p) 2 × 10 − 16 5 × 10 − 9 Note: ∗ p<0.1; ∗∗ p<0.05; ∗∗∗ p<0.01 The instrument for (2) is Tax Admin and the instrument for (3) is License. Standard errors are heteroskedasticity-consistent (HC3). The numbers in round brackets () are standard errors and p means pvalue. 22The Journal of Entrepreneurial Finance, Vol. 27, Iss. 1 [2025], Art. 1 https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 DOI: 10.57229/2373-1761.1505 Table 10: Effects of Degrees of Corruption on Financial Constraints (All Firms) Dependent variable: Financial Constraints IV IV (1) (2) Corruption (Minor) 1.376 ∗∗∗ 1.711 ∗ (0.350) (0.889) Corruption (Moderate) 1.100 ∗∗∗ 0.616 (0.408) (0.387) Corruption (Major) 0.880 ∗∗ 1.539 ∗∗ (0.405) (0.629) Corruption (Very Severe) 1.157 ∗∗∗ 1.262 ∗∗∗ (0.254) (0.444) MSME 0.213 ∗∗∗ 0.219 ∗∗∗ (0.065) (0.081) Capital City 0.152 ∗∗∗ 0.184 ∗∗ (0.039) (0.075) Business City − 0.141 ∗∗∗ − 0.168 ∗∗∗ (0.032) (0.058) South − 0.056 0.018 (0.053) (0.078) Manufacturing 0.025 0.021 (0.032) (0.042) Retail 0.058 0.058 (0.039) (0.052) Firm Age 0.004 ∗∗∗ 0.005 ∗∗ (0.001) (0.002) Revenue 0.001 0.002 (0.006) (0.008) Innovation 0.033 0.034 (0.031) (0.045) Sole Proprietor − 0.021 0.008 (0.037) (0.047) Manager Education − 0.011 ∗∗ − 0.013 ∗ (0.005) (0.007) Female Owner 0.042 0.003 (0.043) (0.053) Subsidiary Firm − 0.026 − 0.028 (0.031) (0.040) Constant − 0.256 − 0.515 (0.202) (0.394) Observations 1,806 1,806 Weak IV (Minor) (p) 4 × 10 − 10 4 × 10 − 8 Weak IV (Moderate) (p) 6 × 10 − 9 6 × 10 − 7 Weak IV (Major) (p) 7 × 10 − 11 5 × 10 − 7 Weak IV (Very Severe) (p) 2 × 10 − 16 2 × 10 − 16 Wu-Hausman (p) 1 × 10 − 14 3 × 10 − 7 Note: ∗ p<0.1; ∗∗ p<0.05; ∗∗∗ p<0.01 The instrument for (1) is Tax Admin and the instrument for (2) is License. Corruption (Minor) was estimated using Tax Admin (Minor) and License (Minor) as instruments and so on. The numbers in round brackets () are standard errors and p means p-value. Standard errors are heteroskedasticity-consistent (HC3). 23 Ezeibekwe: Bureaucratic Corruption and Firms' Financial ConstraintsPublished by Pepperdine Digital Commons, 2025 Table A.1: VIF of OLS Result VIF Corruption 1.026 Capital City 1.354 Business City 1.310 South 1.196 Manufacturing 1.401 Retail 1.378 Firm Age 1.085 Revenue 1.079 Innovation 1.044 Sole Proprietor 1.048 Manager Education 1.106 Female Owner 1.054 Subsidiary Firm 1.164 Table A.2: Summary Statistics for All Firms Statistic Mean St. Dev. Min Max N Corruption (Minor) 0.215 0.411 0 1 1,806 Corruption (Moderate) 0.184 0.387 0 1 1,806 Corruption (Major) 0.337 0.473 0 1 1,806 Corruption (Very Severe) 0.127 0.333 0 1 1,806 Corruption (No Obstacle) 0.137 0.344 0 1 1,806 Tax Admin (Minor) 0.375 0.484 0 1 1,806 Tax Admin (Moderate) 0.267 0.443 0 1 1,806 Tax Admin (Major) 0.162 0.368 0 1 1,806 Tax Admin (Very Severe) 0.029 0.169 0 1 1,806 Tax Admin (No Obstacle) 0.167 0.373 0 1 1,806 License (Minor) 0.378 0.485 0 1 1,806 License (Moderate) 0.237 0.425 0 1 1,806 License (Major) 0.095 0.293 0 1 1,806 License (Very Severe) 0.015 0.121 0 1 1,806 License (No Obstacle) 0.275 0.447 0 1 1,806 Table A.3: Summary Statistics for Recoded Variables Statistic Mean St. Dev. Min Max N Financial Constraints1 0.485 0.500 0 1 1,696 Corruption1 0.654 0.476 0 1 1,696 Tax Admin1 0.461 0.499 0 1 1,696 License1 0.349 0.477 0 1 1,696 Financial Constraints2 0.076 0.265 0 1 1,696 Corruption2 0.127 0.333 0 1 1,696 Tax Admin2 0.031 0.172 0 1 1,696 License2 0.016 0.125 0 1 1,696 30The Journal of Entrepreneurial Finance, Vol. 27, Iss. 1 [2025], Art. 1 https://digitalcommons.pepperdine.edu/jef/vol27/iss1/1 DOI: 10.57229/2373-1761.1505