| چکیده انگلیسی مقاله |
Purpose: Connected lending, defined as the provision of loans by banks to their related parties, represents a critical phenomenon in financial systems worldwide with implications for resource allocation efficiency, financial stability, and governance. This practice can stem from either of two primary mechanisms including a) an informational advantage, where ownership ties reduce information asymmetry and enhance lending efficiency or b) a looting perspective, where banks channel funds to their related entities to the detriment of depositors and minor shareholders. In the context of Iran, where bank-firm ownership relationships are pervasive and the banking sector plays a dominant role in corporate financing, the prevalence and nature of connected lending remain underexplored. Despite the strong interconnectedness between banks and firms, coupled with a regulatory framework that defines the related parties, empirical evidence on how ownership ties influence loan supply decisions is scarce.
This study seeks to address this gap by examining the existence and magnitude of the mechanisms underlying the connected lendings in Iran’s banking system, focusing on the non-financial firms registered over the period from 2007 to 2018. The research aims to quantify the causal impact of bank-firm ownership connections on loan allocation and to discern whether such lending aligns with an informational or looting framework. Thus, it contributes to the literature on financial intermediation and offers insights into the governance challenges facing emerging economies with concentrated ownership structures.
The significance of this investigation lies in its potential to illuminate the allocation of banking resources in Iran, a market characterized by severe financial constraints and a heavy reliance on bank credits due to an underdeveloped corporate bond market. Moreover, the widespread ownership ties between banks and firms (47% of the firms observed in this study) raise concerns about the impartiality of credit allocation.
This study posits two central hypotheses. First, the ownership ties between banks and firms increase the likelihood and amount of loan provision. Second, the connected lending in Iran predominantly reflects a looting mechanism, where riskier related firms disproportionately benefit from credit access.
These hypotheses are tested using a unique hand-collected dataset that integrates bank and firm-level data, providing a granular view of the lending dynamics in an understudied economic context.
Methodology: To investigate connected lending in Iran, this study constructs a novel dataset by merging four distinct sources including a) financial statements of the non-financial firms in the stock market, b) stock portfolios of firms, c) stock portfolios of banks, and d) loan portfolios of banks. This dataset encompasses 260 registered manufacturing firms and 31 banks over the period from 2007 to 2018, yielding 54549 bank-firm-year observations after data cleaning. The ownership relationships are meticulously hand-collected and categorized into five levels of complexity, ranging from direct bank ownership in a firm to indirect ties through multiple intermediaries. Also, cash flow rights and voting rights serve as two measures of ownership.
The empirical strategy leverages a regression framework to estimate the causal effect of ownership ties on loan supply, controlling for the confounding factors. A key innovation is the use of firm-specific credit demand shocks, constructed following Amiti and Weinstein (2018), to disentangle supply-side effects from demand-side influences. This approach isolates the impact of ownership connections on the lending decisions of banks by accounting for the borrowing needs of firms. Additionally, prior bank-firm lending relationships are controlled to distinguish the effect of ownership from relational lending dynamics.
The dependent variables include a) a dummy for loan receipt, b) a dummy for a bank entering the loan portfolio of a firm, c) the share of the bank in the loan portfolio of the firm, and d) the loan-to-sales ratio. These variables capture different dimensions of lending behavior, from probability to volume. Robustness is assessed by varying the ownership definitions and thresholds.
Findings and Discussion: The results provide robust evidence for connected lending in Iran’s banking system. Ownership ties significantly increase both the likelihood and magnitude of loan provision. Specifically, establishing a bank-firm ownership relationship raises the probability of loan reception by 2.11 percentage points and the likelihood of a bank entering the loan portfolio of a firm by 2.06 ppts. Furthermore, such connections elevate the share of the bank in the loan portfolio of the firm by 0.9 ppts and the loan-to-sales ratio by 33%, compared to the sample mean. These effects are economically substantial, given the baseline probabilities and loan volumes in the data.
The relationship between ownership stakes and lending is non-linear, with a diminishing marginal effect as ownership increases. This suggests a trade-off between the benefits granted to the related firms and the costs incurred by the bank stakeholders. A critical finding is that riskier connected firms, i.e., those with higher leverage, greater financial debt, or lower cash reserves, receive disproportionately more loans. This pattern aligns with the looting hypothesis, where banks prioritize related parties regardless of creditworthiness, potentially at the expense of depositors and minor shareholders.
The results are robust to alternative ownership measures (voting rights) and varying thresholds, including those below the 10% related-party definition of the Central Bank of Iran’s. This indicates that connected lending occurs even at low ownership levels.
Conclusions and Policy Implications: This study confirms the pervasive existence of connected lending in Iran, highlighting its significant role in credit allocation. The findings underscore a looting-driven mechanism. The non-linear effect of ownership suggests a strategic calculus by bank insiders, balancing gains in related firms against losses in the bank value. The persistence of connected lending below regulatory thresholds points to deficient oversight, as firms exploit complex ownership structures to circumvent Central Bank rules.
These insights carry important policy implications. First, the Central Bank should enhance transparency in related-party transactions, as lowering ownership thresholds alone may be insufficient given the intricate networks involved. Second, the preferential treatment of riskier firms signals a need for stricter governance and risk management protocols to safeguard the banking system’s stability.
While this study focuses on the firms registered in the stock market, extending the analysis to non-listed entities, potentially less transparent, could reveal even greater distortions. Similarly, exploring how ownership network structures influence lending patterns offers a promising avenue for future research. In conclusion, this research not only advances our understanding of connected lending in Iran but also underscores the broader challenges of governance and financial intermediation in economies with concentrated ownership and limited market discipline. |