Amir Sufi (2007) carried out an empirical research to examine how information asymmetry influence the structure of financial institutions that participate in a syndicated loan arrangement, in view of Holmstrom 1979 theoretical model of agency and moral hazard. He also examines how lead arrangers of syndicated loans structure the syndicates to reduce the moral hazard problem. The fundamental assumptions about Holmstrom and the Tirole is that, firm's in which the public can scarcely get information will require an informed lender to monitor and perform due diligence before other lender can be participant of the syndicate. In this paper, I review Sufi's research paper with a sample size of 12,672 syndicated loan deals in United State from 1992- 2003. I examine how Participant lenders and borrowers attributes would control or increase the problem of information asymmetry between lenders or borrowers and the effect on the syndicate loan structure for obscure or public borrowers. In other to do this, I divide the report into different sections. Section A borders on the theoretical and empirical test of sufi's paper, section B examine the research methodology, C dwells on the findings and the final section is the conclusion of the report.
Theoretical frame work and Empirical Assumptions on which Sufi's Paper is based
The key aspects of macroeconomic instability that predicts the information asymmetry between firms and their investors as suggested by Bernanke and Eertler (1989), Holmstrom and tirole (1997) are agency problem and constraints on external financing. Sufi (2007) pointed out that models on information asymmetry that has been used in corporate finance, which includes Daimond (1991) and Boot and Thakor (2000) empirically explains the difference that exist between loans that are influence by relationship banking and public debt. He asserts that syndicated loans can be positioned in between the aforementioned extremes, since it has similar characteristics of private loans and public debt issues which are backed or underwritten by a financial institution. The frame work used by sufi is consistent with early banking theories which suggested that information asymmetry and the need for monitoring are very key in the banking operations (Lel and Pyle 1997). The lead arranger is an informed lender who is also able to monitor and learn about the borrowing firm through unobservable effort that might be very costly. However, there would be a problem of moral hazard with the lead arranger since his effort is not directly observable by the potential participant lenders.
B. METHODOLOGY
Sufi (2007) examines how information asymmetry influences the structure of a syndicated loan. In other to prove the aforementioned null hypothesis, he performed several test using 4,689 syndicated deals in the United States. Using a general regression model defined as
Syndi = α + t+ Xiβ + opaque γ + εi ----------------------------------------- (1)
The variable in the left hand of the regression equation is the dependent variable, while the right hand variables represent the explanatory variables. Syndi measures the numbers of lead arranger that participated in the syndicated loans and the proportion retained by the informed arranger. Opaque measures the extent at which the financial institutions should investigate and monitor borrowers that are very difficult to value, γ is the coefficient of interest that explains the structure of the syndicate. X is the control variables, which includes the natural log of firm's sales, industrial indicators variables and other loans characteristics. Dumt is the dummy variables at time t.
The criteria for chosen participant of the syndicates was also analysed by examining the probability of a lender being chosen in a particular syndicate deal given certain factors. This was done using an utmost probability choice model which is consistent with the model used by Corwin and Schutz(2005) to explain the choice of IPO syndicate members. The participant's choice consists of all financial institutions that represent 0.5% of syndicated loan participants in the year of the loan in question. The probit model is generalised by the following equation
Pr(Participant= Bankij) = f(α + β - Loani + γ - Bankj + εij) ------------------------(2)
The generalised probit model seek to explain how the characteristics of loan i bank j influence the probability that bank j is selected as a participant on the loan i. While γ is the parameter of interest that changes in line with the opacity of the borrowing firm.
The information asymmetry impacts on the syndicate loans structure is subdivided into the following test inference and findings.
FINDINGS
C1. Discrepancy in the opacity of borrowing firm's influence the structure of the syndicate. The result supports the theoretical frame work of information asymmetry on agency and moral hazard among the syndicates. When borrowing firms are very difficult to value, in other words they don't have publicly available information like Sec filings. The participant lenders in the syndicated loan would depend solely on the lead arranger for comprehensive information on the borrowing firms. The absence of Sec controls over the borrowing firm would mean, the penalty for miss-informing the public about their financial position would not be severe. Therefore firms that are not overseen by Sec would require additional due diligence and monitory. The point estimate also gives a result that support moral hazard in an arrangement of information asymmetry. The lead arranger holds the largest proportion of the loan and most concerted syndicate with fewer participants when the borrowing firms are private. This pattern is also consistent with borrowing firms that are public but unrated. However, it is with a lesser degree compared to when the firms are public.
C2. Borrowing firm's reputation affect the syndicate structure. In this inference, Sofi assumed that the syndicated market involves repeatative interactions between the lead arrangers and the borrowing firms. And so, the borrowers should become better known to the participant banks the more time they assess the syndicated market for loan. This invariably implies the borrowers assumed the lead bank has done his due diligence. In this case the lead bank should lend a lower portion of the syndicated loan when the borrowing firms are familiar borrowers as predicted by the information asymmetry frame work. Sofi's result supports this position. He grouped both and unrated and private borrowers into one group through a cross sectional regression and analyse how the syndicate structure changes when borrowing firms repeatedly asses the syndicate market. He found out that lead arranger hold large proportion of the loan when the borrowing firm is asset is difficult to value(0paque) and less proportion of the loan when the borrowing firm has previously assed the market.
C3. The lead bank reputation influence the amount of the syndicated loan held by him. According to sufi, another answer to moral hazard ascribed to the lead arranger is reputation. He measures reputation of lead arranger by market share of syndicated loans in previous year. And found out that only reputable lead arrangers in the extreme right tail of the normal distribution can fully mitigate information asymmetry with reputation. In other words its can only happen in rare cases. However, his findings shows that the effect of lead arrangers reputation to substitute for lead arrangers satisfying other participating banks fears on due diligence is more consistent with opaque firms.
C4. Does moral hazard explains the proportion of syndicated loan held by the lead arranger better than adverse selection. The adverse selection implies that the large portion of the loan lend by the lead bank is to signal to the participant lenders that he is not selling a bad loan to them. This is because if a lead arranger has hidden information on the borrowing firm that is not available to the participant lenders, he may be tempted to give out large portion of the loan when the information he has about the borrowing firm is depressing. While the moral hazard implies that the participant lenders sees the large portion lend by the lead bank as a stimulant to compel him to perform is monitoring and due diligence on the borrowing firm thereby mitigating against any default. To prove these frame work Sufi tested two hypotheses and found out that it is only the moral hazard predictions that holds. Which suggest that a lead arranger with previous lending relationship with the borrowing firm has already performed his due diligence and should hold less of the loan in the syndicate.
C5. Does a well known borrower in the syndicated loan market influence the syndicate's structure? In the regressed probit model analysis, Sufi found out that there was a significant persistent in the borrowing firm and participant relationship. There is also about 27.3% likelihood of a participant lender that has dealt with a borrowing firm to be chosen in a current deal. However, these factors are relatively more pronounced if the borrowing firm is either a private or unrated firm. The overall result also suggests that previous borrowing firm and participant relationship is more important than previous lead bank and participant relation. This implies that to reduce the effect of information asymmetry the lead bank would prefer to choose participants that are familiar to the borrowing firm rather than those close to him.
CONCLUSION
This paper reviews how syndicated loans have been playing a major role in corporate financing. It also shows that financial institutions are more concerned about retrieving their credit facilities rather than interbank relationship in the financial market. The paper also examines the evidence that information asymmetry controls the structure of a syndicate loan, and that the reputation of a lead bank is very key in reducing the impact of information asymmetry.
References List
Ben, Bernanke, and Mark Gertler. (1998). Agency costs, net worth and business fluctuations, American Economic Review 79, 14-31
Boot, Arnoud, and Anjan Thakor, (2000). Can relationship banking survive competition? Journal of Finance 54, 679-713.
Corwin, Shane A., and Paul Schultz, (2005). The role of IPO underwriting syndicates: Pricing, information production, and underwriting completion, Journal of finance 60, 443-486
Diamond, Douglas. (1991). Monitoring and reputation: The choice between bank loans and privately placed debt, Journal of Political Economy 99, 689-721
Holmstrom, Bengt. (1979). Moral hazard and the observabilty, Bell Journal of Economic Review 10, 74-91
Holmstrom, Bengt, and Jean Tirole. (1997). Financial Intermidiation, loanable funds, and the real sector, Quarterly Journal of Economics 112, 663-691
Leland, H., and D. Pyle.( 1997). Informational asymmetricies, financial structure, and financial intermediation, Journal of Finance 32, 371-387.
