Do Firms Benefit from Having a Fortune 500 Customer/Supplier? Evidence from the Loan Market

Transcription

1 Do Firms Benefit from Having a Fortune 500 Customer/Supplier? Evidence from the Loan Market Hsiangping Tsai College of Management Yuan Ze University 135 Yuan-Tung Road, Chung-Li, Taiwan Phone: ext Fax: hptsai@saturn.yzu.edu.tw Chuan-Ying Yang College of Management Yuan Ze University 135 Yuan-Tung Road, Chung-Li, Taiwan Phone: ext Fax: s997404@mail.yzu.edu.tw Haoyi Wang College of Management Yuan Ze University 135 Yuan-Tung Road, Chung-Li, Taiwan Phone: ext Fax: s989503@mail.yzu.edu.tw JEL classification: G15; G20; G21 Keywords: Fortune500, Supply Chain relationship, Terms of Loan 1

2 Do Firms Benefit from Having a Fortune 500 Customer/Supplier? Evidence from the Loan Market ABSTRACT This study examines whether a firm s customer-supplier relationship affects its loan terms. Large and well-performed firms, such as Fortune 500, are prudent in selecting their supply chain partners. From bank lenders view point, such relationship can convey information regarding the borrower s quality. Thus, we infer that a firm with a Fortune 500 as main partner in the supply chain is of good quality. Consistent with this idea, we show that borrowers with a Fortune 500 main partner in the supply chain obtain loans with lower spread and larger loan amount. We further examine whether firms benefit from the Fortune partners trade credit extension during crisis periods. The results show that, during crisis period, borrowers themselves as Fortune suppliers (customers) have significantly larger accounts receivable (lower accounts payable), while borrowers with Fortune customers have significantly lower accounts receivable, indicating that Fortune firms help their partners through the trade credit arrangements. That is, during crisis periods, Fortune firms extend trade credit and speedup payments to their partners. Keywords: Fortune 500, Supply Chain, Syndicated Loans 2

3 1. Introduction Existing studies have shown that a firm s customer-supplier relationship affects its performance, corporate policies, and shareholder wealth. Some studies examine the effect of a firm s specific event on its suppliers, customers, and rivals, such as merger and acquisitions (Fee and Thomas 2004; Shahrur 2005), and bankruptcy filings (Hertzel et al. 2008). Others analyze how the customer-supplier relationship affects a firm s capital structure choice (Kale and Shahrur, 2007), information disclosure decision (Ellis, Fee and Thomas 2012) and investor wealth (Pandit, Wasley, and Zach 2011; Wang 2012). Results of these studies imply that a firm s customer-supplier relationship contains crucial information about its position, competitive relations in the product market, and future prospects in terms of performance. We expect that such relationship could also affect the firm s financing activities, i.e., the interaction between a firm and its bank lenders. In this study, we aim at investigating whether bank lenders rely on such relationship to gather information about its borrowers. Literature regarding the bank-firm relationship indicates bank s role in generating information is the key to mitigate the information asymmetry between the two parties and thus both parties can benefit from maintaining good relationship with each other. To name a few, with expertise in screening and interaction with the borrowers, banks generate valuable information about the borrowers and hence bank loans serve a certification role to signal borrower quality to outside investors (James 1987, Lummer and McCollell 1989, Slovin, Sushka and Hudson 1990; Dahiya, Puri, and Saunders 2003) and banks benefit from higher likelihood of obtaining future business from existing customers (Bharath et al. 2007). With good banking relationship, benefits to borrowers includes increased funding availability for small firms with close ties to banks (Petersen and Rajan 1994), better performance during crisis period for firms 1

4 with high-quality banks (Fok, Chang, and Lee, 2004), and smaller spreads, larger amounts and reduced requirements on collaterals for firms that repeatedly borrow from the same lender (Bharath et al. 2011). In this study, we propose that banks can rely on the customer-supplier relationship as a source of information to identify a borrower s quality. It is well known that large and well-performed firms, such as Fortune 500, are prudent in selecting their supply chain partners. If a firm is selected as a main supplier of a Fortune 500, such relationship can convey information regarding the non-fortune firm s quality to its lender. Thus, we expect lenders may incorporate the information into the decisions of loan terms by assuming that a firm s main partner in the supply chain is a Fortune 500 as good quality. Consistent with this idea, we show that borrowers with a Fortune 500 main partner in the supply chain obtain loans with lower spread and larger loan amount. We further examine whether firms benefit from the Fortune partners trade credit extension during crisis periods. The results show that, during crisis period, borrowers themselves as Fortune suppliers (customers) have significantly larger accounts receivable (lower accounts payable), while borrowers with Fortune customers have significantly lower accounts receivable, indicating that Fortune firms would extend trade credit or speedup payments to help their suppliers/customers during crisis periods. The rest of this paper is organized as follows. Section 2 provide theoretical background and develop the empirical hypothesis. Section 3 describes the data and my sample. In Section 4, we present the empirical design and variable definitions in the model. Section 5 reports the empirical results. The final section concludes. 2

5 2. Related literature and hypotheses 2.1 The importance of supply chain Previous studies have shown that the supply chain plays an important role to affect firms along the chain in different financial dimensions. Some scholars use firm specific events to study the impacts on firms along the supply chain. Fee and Thomas (2004) examines the effects of horizontal merge and acquisitions on rivals, customers and suppliers to identify the source of gains to horizontal M&As. Their evidence supports the improvements in efficiency of production and purchasing power, but does not support for collusions with rivals. In support of efficiency considerations for takeovers, Shahrur (2005) demonstrates the positive wealth effect of horizontal takeovers is associated with significant positive abnormal returns to rivals, suppliers and customers. On the other hand, Hertael et al. (2008) examine the impact of a firm s financial distress along the supply chain. The effects on suppliers/customers depend on the industry structure, i.e., whether there is a contagion effect on the rivals in the bankruptcy firm s industry. If the rivals experience contagion effects, suppliers suffer more from the contagion, but customers are less affected. Kale and Shahrur (2007) investigate the impact of supply chain relationship on a firm s capital structure decision. They propose that firms would reduce leverage (i.e. to keep low likelihood of liquidation) to convince their suppliers/customers to develop relationship-specific investments. They provide evidence that a firm s leverage is more likely to be low when R&D intensities of suppliers/customers are high, when strategic alliances and joint ventures along the supply chain are more common, and when the industry of supplier/customer is more concentrated. 3

6 Some other scholars study the interrelationship between a firm s information disclosure and reactions of investor along the supply chain. Ellis, Fee and Thomas (2012) show that firms avid disclosing proprietary information about customers when they concern that such action may turn out to be aiding rivals. Pandit, Wasley and Zach (2011) documents information externalities along the supply chain by showing that a firm s stock price reacts to the earnings announcements of its major customers. Existing study shows that the interaction among firms along the supply chain contains crucial information about a firm s business relation, position in the product market, and its future prospects in terms of performance. Such crucial information should also affect a firm s financing activities. However, we know little about this field. Only a few studies try to link a firm s supply chain relation to its financing activities. For example, Files and Gurun (2014) document evidence that a firm s loan terms are negatively affected by the financial restatement announced by peer firms and firms along the supply chain. 2.2 Bank s role in generating information In the process of loan making, banks play a key role in generating borrower information to mitigate the information asymmetry problem and thus both parties can benefit from maintaining good relationship with each other. It is well known that bank loans serve a certification role to signal borrower quality to outside investors (James 1987; Lummer and McCollell 1989; Slovin, Sushka, and Hudson 1990). In terms of benefits to borrowers, Petersen and Rajan (1994) documents small firms that buildup close ties with a lender benefit from increased funding availability. Bharath et al. (2011) offers evidence that firms may acquire better loan terms by repeatedly borrowing from the same lender, including larger loans and reduced requirements on collaterals. For less transparent borrowers, 4

7 borrowing from a relationship lender who is familiar with the borrower further helps reduce their loan spread. In terms of benefits to banks, Bharath et al. (2007) provides evidence that the probability of relationship banks to obtain future business from corporate customers is significantly higher than that of non-relationship banks. Evidence regarding lenders effort in gathering information to help monitor the borrowers is also well-documented. For example, if the borrower is information problematic, lead arranges would hold a larger loan share, arrange a more concentrated syndicate, and invite participant lenders who are more familiar with the borrower (Dennis and Mullineaux 2000; Lee and Mullineaux 2004; Sufi 2007). In addition, when borrowers are in financial distress, lenders would suffer significant losses (Dahiya, Saunders and Srinivasan 2003). 2.3 Bank-firm relationship and access to credit during crisis period One strand of literature focuses on exploring the effects of financial crisis or bank distress on the bank-firm relationship. Slovin, Sushka, and Polonchek (1993) are the first to document evidence that the failure of a bank harms its borrowers. Their evidence indicates the insolvency of Continential Illinois Bank had negative impact on its borrower stock prices. Fok, Change, and Lee (2004) examine the impact of bank relationship on firm performance for Taiwanese firms during 1997 Asian financial crisis. They show that borrowing firms performed better around Asian financial crisis if they were with high-quality banks or foreign banks. Looking at U.S. firms during the Russian crisis period, Chava and Purnanadam (2011) offers evidence that bank-dependent firms suffered larger declines in value, capital expenditure, and performance than firms that were less-dependent on banks. Above findings suggest that a firm s funding source is restricted when its lender experiences 5

8 unanticipated shocks and thus the performance will be negatively affected. Some scholars suggest that firms with limited access to credit market may use trade credit from suppliers as a source of funds. For example, Petersen and Rajan (1997) suggest that suppliers have advantage in gathering information about buyers and thus are able to offer trade credit to firms that are constrained from bank financing. Cull, Xu, and Zhu (2009) show that China firms with limited access to formal bank credit tended to use trade credit from profitable suppliers as a substitute for loans. Love, Preve, and Sarria-Allende (2007) examine effects of show that firms that are more vulnerable to financial crisis reduce the provision of trade credit to customers. In addition, Giannetti, Burkart and Ellingsen (2011) suggest that the extension of trade credit conveys favorable information to uninformed lenders, which become more willing to lend. Based on existing studies, our analysis includes trade credit and assumes that Fortune 500 firms are less vulnerable to financial crisis and are more likely to extend trade credit. 2.4 Hypotheses This study aims at linking the two strands of literature to explore the effect of supplier-customer relationship on a firm s loan characteristics. Specifically, for each borrower, we classify its main business partners along the supply chain into Fortune 500 firms or others. Since the Fortune 500 tend to have stringent requirements in selecting their corporate partners, we believe that a firm has a Fortune 500 firm in its supply chain help improve its operations, performance, and thus its overall quality. We expect that bank lenders may take such firms as of good quality (i.e, lower likelihood of default) and will be more willing to offer better loan terms to them. Hypothesis 1: A firm with a Fortune 500 supplier/customer is more likely to obtain better loan 6

9 terms. Also, we consider the supplier-customer relationship to have different effect on firm s funding source during crisis period. The benefit of being a partner of a Fortune 500 could be that a firm would receive better trade credit terms from its Fortune partner during the crisis periods. Thus, we create dummy variables for crisis periods to examine the effect of the supplier-customer relationship on a firm s trade credit in different economic cycles. Hypothesis 2: A firm with a Fortune 500 supplier/customer is more likely to receive better trade credit terms from its Fortune partner during crisis periods. 3. Data and sample The relationship between suppliers and customers is collected from Compustat Segment data. Compustat Segment data reports a firm s main corporate customers. For each firm, we identify a unique supplier-customer relationship. A firm s major customer is the corporate customer that accounts for the highest proportion of its sales revenue. A firm s major supplier is the supplying firm that accounts for the highest proportion of its cost of goods sold. In other words, each firm has a most important customer/supplier. Then, we obtain the Fortune 500 rank from Compustat and match it to our supplier-customer sample. Based on the Fortune 500 information, we are able to identify 3 possible combinations. First, both the supplier and customer are not Fortune 500. Second, either the supplier or the customer is a Fortune 500. Third, both the supplier and the customer are Fortune 500. Loan data comes from Dealscan loan database. We use Dealscan-Compustat link from 7

10 Chava and Roberts (2008) to connect the supplier-customer sample with the loan data. For each loan, we rely on the information during 3 years prior to the deal active date to identify whether the borrowing firm has a Fortune 500 supplier/customer or not. Lastly, we obtain borrower s financial information from Compustat database. The final sample contains 1,512 loan deals for the supplier firms and 2,411 loan deals for the customer firms from 1984 to Table 1 exhibits the descriptive statistics of our sample. Panel A in Table 1 provides the information about the number of supplier firms and customer firms in syndicated loan market by year. It shows from 1996 to 2007 the syndicated loan market boomed. Both for supplier and customer, there are more than 4% of firms joining in syndicated loan market every year during this period. Panel A in Table 1 also displays the distribution of the quantity of syndicated loan deals. There are 1,512 loan deals made to suppliers and 2,411 loan deals made to customers during the sample period. Then, it is consistent with the left-hand side of Panel A that the syndicated loan market is prosperous from 1996 to According to the number of firms and number of deals, the peak of syndicated loan is from 2000 to 2004, which include over 7% of firms and deals. The number of loan deals is obviously much larger than other time period. From Panel B of Table 1, we can observe how many supplier firms and customer firms in the sample and how many of them had ever been included in the Fortune 500 list. There are 742 supplier firms and 722 customer firms in our sample. It shows that 36% of the customers had ever been included in Fortune 500, but only 5% of suppliers had ever been included in Fortune 500. Fortune 500 lists contain more companies which are customers in supply chain such as Wal-Mart, so there is a quite asymmetrical distribution on supplier and customer in the sample. [Insert Table 1 here] 8

11 4. Methodology The first issue of this paper is whether the firms with Fortune 500 supplier/customer can get better loan terms when they obtain funds from loan market. Secondly, we also investigate trade credit of firms during different economic situations. For this reason, we develop the following models to test our hypothesis. Please refer to Table 2 for detail information about variable definitions. Term it β 1 Fortune_partner it β 2 Fortune_borrower it β 3 Fortune_both it β 4 Crisis it n j 5 β j X it,j ε it (1) Trade Credit it β 1 Fortune_partner it β 2 Fortune_borrower it β 3 Fortune_both it β 4 Crisis it 7 n j 5 β j *FortuneDummy it,j k 8 β k X it,k ε it (2) 4.1 The terms of loans Term is the terms of loans made to borrowers. In this study, we examine loan amount, loan spread and the maturity. Generally speaking, the high quality firms are more likely to acquire larger amount of loans from banks. We use deal amount (DealAmt) to measure the loan amount and the unit of variable is millions of dollar. On the other hand, firms with good quality are more likely to obtain loans with lower spread, and vice versa. We apply all-in spread drawn (AllInDrawn) to measure the loan spread, which is the total costs borrowers pay for their loan and contains coupon spread plus annual fee. Maturity (Maturity) represents the extension of loans 9

12 in number of months. 4.2 Fortune 500 as a mark of high quality firms Based on publicly available revenue data, Fortune Magazine publishes a list for the most 500 successful U.S. corporations as Fortune 500 in each year. Being included as Fortune 500 is considered as a mark of prestige, i.e., high quality and low default probability from a banker s view point. Thus, most bankers are willing to offer better loan terms to such firms. In this study, we hypothesize that a firm may benefit from having a high quality firm in its supply chain when it raises bank loans. The possible benefits could be the availability or cost of funds. Therefore, we rely on Fortune 500 firms to measure the quality of a firm s partners in its supply chain. In order to examine the above benefits, we set three dummy variables into the model. If both the supplier and the customer are not Fortune 500 firms, it will be the basic group of the sample. Fortune_partner is one if a borrower is not a Fortune 500, but its partner (supplier/customer) is a Fortune 500; zero otherwise. Fortune_borrower is one if the borrower itself is a Fortune 500, but its partner (supplier/customer) is not a Fortune 500; zero otherwise. Fortune_both is one if both the borrower and its partner (supplier/customer) are in the Fortune 500; zero otherwise. We are curious about whether banks would base on the supply chain relationship to determine the terms of loans. So, we use DealActiveDate in DealScan loan database from Thomson Reuters LPC for further analysis. Based on the DealActiveDate, we search whether borrowers have Fortune 500 suppliers/customers in the previous three years. We assume that banks would depend on past supply chain relationship to determine the current loan deals. If a 10

13 firm with at least a Fortune 500 supplier/customer in the previous three years, we identify it as a firm with Fortune 500 business partner. 4.3 Crisis period Previous research has shown that banks may reduce lending and raise the loan interest rates during crisis period, thus we also consider the crisis effect on the loan terms. We indentify three substantial crisis periods as bad time, including Asian crisis (1997/07/02~1998/12/31), dot-com crisis (2000/03/10~2001/12/31) and 2008 Financial crisis (2008/09/15~2009/06/30), respectively. 1 Based on these crisis periods, we set a dummy variable - Crisis to distinguish the good time and the bad time. Crisis equals to one if a loan was arranged in crisis period, otherwise Crisis equals to zero. 4.4 The use of trade credit during crisis periods The fact that banks tend to reduce lending in crisis periods may have relatively smaller effect on high quality firms, such as Fortune 500. But, such reduction in lending could have significantly negative effect on other firms. Therefore, we ask whether the supplier-customer relationship helps to alleviate such negative effect by allowing firms to rely on trade credit during an economic downturn, i.e. financing based on the supplier-customer relationship. For suppliers, we use TC_AR to measure the trade credit effect. TC_AR is account receivables divided by total assets. For customers, we use TC_AP to measure the trade credit effect. TC_AP is account payables divided by total assets. To supplier borrowers, increased accounts receivables indicate extending trade credit to customers, while reduced accounts receivables imply collecting 1 The time periods of these three substantial crises are identified by some news on internet. The news provides the complete process of those crises. The beginning date of crisis is the day that crisis exploded. Then, based on news, I can realize how long the crisis it sustained. The end day of crisis is the fiscal year end date that one or one and half year later the crisis. 11

14 receivables more quickly. To customer borrowers, reduced accounts payables indicates speeding up payments, while increased accounts payables imply delaying payments. With good supplier-customer relationship, we expect Fortune 500 firms would extend trade credit or speedup payments to help their suppliers/customers during crisis periods. 4.5 Other control variables Size. We use a firm s total assets (TA) to measure its firm size. In comparison to small firms, large firms have more resources, more stable performance, and are less likely to default. Thus, they are more likely to obtain better loan terms. Investment. PP&E (PPE) measure the fixed assets of firms for their business operations. In the process of loan offering, a firm with high PP&E indicates the lender could take it as a source of repayment. We use PP&E scaled by total assets to mitigate the size effect on PP&E. Tobin s q (TBQ) is the book value of debt plus the market value of equity divided by total assets. A higher Tobin s q indicates a firm has better investment opportunities. Leverage. Cash (CASH) measure the firm s ability to repay. As for banks, the cash held by firms directly affects whether borrower will repay their debt on schedule. The firms with more cash holdings have higher probability to pay back their loans. We also scale the cash of firms by their total assets in order to eliminate the size effect as PP&E case. Leverage (LVG) is total debt divided by total assets. Firms with high leverage are more likely to default than firms with low leverage. Thus, other things being equal, high leverage firms are less likely to obtain better loan terms from banks. Profitability. We use nature log of sales (LNS) and earnings before interest, tax, depreciation and amortization (EBITDA) to evaluate a firm s business performance and profitability. EBITDA 12

15 is scaled by total assets. A firm with higher profitability is less likely to default and thus can obtain better loan terms. Credit quality. We obtain S&P credit rating from Compustat to represent a firm s credit quality. To transfer the S&P credit rating into rating scores (Rating_score), we assign the value of 1 to AAA rated firms, 2 to AA+, etc. A higher value of Rating_score indicates lower credit quality. 5. Empirical Results 5.1 Firm and loan characteristics and by borrower type Table 3 presents firm characteristics by borrower type. We can observe a pattern that, in comparison to supplier firms, a larger proportion of customers in our sample are Fortune 500. From the subsample that suppliers who borrow in the loan market, we can observe 34.4% of their customers are Fortune 500. From the subsample that customers as borrowers, about 48% of themselves are Fortune 500. A further comparison on the firm size (total assets, TA) suggests that firm size of customers (Median=339.5) is much larger than the firm size of supplier (Median=5,033). Both confirm that the sample is somewhat asymmetric as identified in Table 1 Panel B, i.e. more customers are Fortune 500 in our sample. Also, the customer borrowers have significant better credit rating (mean=11.78, median=10) than the supplier borrowers (mean=18.51, median 23). During the sample periods, about 21%~25% of bank loans are offered during crisis period, while more than 75% of loans are offered during normal time. [Insert Table 3 here] In comparison to supplier borrowers, customer borrowers have better profitability in terms 13

16 of sales (LNS), EBITDA (EBITDA), and tax payment (TAX). But, supplier borrowers have higher sales growth (SG). Similarly, supplier borrowers have much larger Tobin s Q (Median=2.590) than customer borrowers (Median=1.806). Focusing on Dividend policy, consistent with the business growth, supplier borrowers have lower dividend payout than customer borrowers, probably due to the need to support for the higher growth. Based on the maturity structure of long-term debt, supplier borrowers seem to have higher long-term debt ratio (LTD1, LTD3 and LTD5) than those customer borrowers. Table 4 presents loan and trade credit characteristics by borrower type, classified by supplier and customer. In general, the characteristics of our customer sample seem to be more diffuse. In comparison to customer loans, on average, loans to suppliers seem to have relatively smaller amount, higher spread, and shorter maturity. Also, suppliers tend to have larger size of accounts receivables and accounts payables, indicating that suppliers rely more on the trade credit. We further classify loans into loans made in crisis versus normal periods. Loans obtained during crisis periods tend to have shorter maturity. For the suppliers, average level of accounts payables tends to be larger during crisis periods, possibly due to the management to postpone payments. [Insert Table 4 here] 5.2 Does having a Fortune 500 supplier/customer help obtain better loan terms? Table 5 presents results for the effect of having a Fortune 500 along the supply chain on a borrower s loan terms. We study the suppliers and customers as borrowers separately and report the results in Panels A and B of Table 4, respectively. [Insert Table 5 here] 14

17 In Panel A Table 5, we examine loan terms with the subsample that suppliers are borrowers. Models (1)-(3) are the results for loan amount (DEALAMT), spread (AllInDrawn), and maturity (Maturity), respectively. When a non-fortune supplier borrower has a Fortune partner, i.e., a Fortune customer, the loan spread is significantly lower (at 10% level), but no effects for loan amount and maturity. When both the supplier borrower and its customer are Fortune 500, the loan spread is significantly lower (at 1% level) and loan maturity is significantly shorter (at 1% level). The results imply that banks may incorporate the supply chain relationship to identify whether a firm is of good quality or not and thus building a good supplier-customer relationship with Fortune Company helps to reduce loan spread. Also, being a Fortune firm is a mark of high quality. Therefore, when both firms along the supply chain are Fortune firms, the reduction in spread is significantly larger. As for the borrower characteristics, most of them are consistent with our expectation. For example, large firms tend to obtain loans with larger amount, lower spread, and longer maturity. Firms with more PP&E are more likely to obtain long-term loans. Firms with better performance (EBITDA), higher growth opportunity (Tobins q, TBQ), and better credit quality (low RATING_SCORE) tend to obtain loans with lower spread. Loans offered during crisis periods (CRISIS) tend to have shorter maturity. In Panel B Table 5, we report the results for the subsample that customers are borrowers. Models (1)-(3) are the results for loan amount (DEALAMT), spread (AllInDrawn), and maturity (Maturity), respectively. When a non-fortune customer has a Fortune partner, i.e., a Fortune supplier, the loan amount is significantly larger (1% level), and the loan spread is significantly lower (5% level). When a customer borrower itself is a Fortune firm but its supplier is not, the loan amount is significantly larger (1% level), and the loan spread is significantly lower (5% 15

18 level). However, when both the borrower and it supplier are Fortune firm, the loan spread become significantly higher (1%), although the loan amount is larger (10% level). If we add up the coefficient for both Forutne_partner and Fortune_both, the net effect of having a Fortune supplier on a customer borrower s spread is still lower than those without a Fortune supplier. The effects of other borrower characteristics are generally consistent with intuition. In general, our results suggest that a Fortune partner may help a firm to obtain better loan terms. The implication to a non-fortune borrower is that maintaining strong relationship with high quality firms, such as Fortune 500, does help to improve the lender s view on its quality. 5.3 Can a firm obtain better trade credit terms from its Fortune partner in crisis periods? In Models (4)-(5), Panel A Table 5, we present results for the effect of a supplier borrower having a Fortune 500 customer on its trade credit, TC_AR, measure by accounts receivables over its total assets. We find that the TC_AR of a supplier who have a Fortune customer, are significantly larger than those without a Fortune customer. However, during crisis period, such firm s TC_AR is significantly lower (1% level). The results indicate that with a Fortune customer, the suppliers extend trade credit to its Fortune supplier in normal time, but can collect receivables more quickly during hard time. Also, when a supplier itself is a Fortune firm, its TC_AR is not different from those who are not Fortune firms. However, during crisis periods, it has significantly lager TC_AR. The results imply that a Fortune supplier tends to extend trade credit to its non-fortune customers during crisis periods. In Models (4)-(5), Panel B Table 5, we study the customer borrower to see the effect of having a Fortune 500 supplier on its trade credit, TC_AP, measured by accounts payables over its total assets. We find when the customer borrower itself is a Fortune 500, it tends to have 16

19 significantly larger TC_AP during normal time, but have significantly smaller TC_AP during crisis periods. The results are consistent with our results in Panel A Table 5 in that Fortune firms speedup their payments to supplier during crisis period, although their payment periods seems to be longer in normal time. Generally, our results are consistent with our expectation that, with strong supplier-customer relationship, Fortune firms would extend trade credit or speedup payments to help their suppliers/customers during crisis periods. 5.4 Further tests The core concept of our study relies on using Fortune 500 to identify whether the largest supplier/customer is of high quality. This design has some drawbacks. For example, a firm may have other important Fortune suppliers/customers, but our analysis excludes them. A firm may have a high quality supplier/customer which is not included in the Fortune 500 list. Thus, we also develop the following criteria to identify the effect of high quality suppliers/customers Customer-supplier relation based on Fortune proportion In this subsection, we measure the customer-supplier relation with the size of the business. We use the proportion of sales to Fortune customers for our supplier borrowers and the proportion of purchase on the cost of goods sold from Fortune suppliers for our customer borrowers. Panel A Table 6 presents results for our supplier borrowers. We identify whether the borrower itself is a Fortune supplier (Fortune_supplier) and the proportion of its sales to Fortune firms (Fortune_CR). Models (1)-(3) reports results for loan terms. As expected, loan spread is 17

20 significantly lower when the borrower itself is a Fortune firm and when the borrower has larger proportion of sales goes to Fortune firms. In addition, we also find that a Fortune supplier tends to obtain loans with shorter maturity. In Models (4)-(5), Panel A Table 6, we also conduct this analysis for the trade credit terms, but do not obtain significant Fortune firm effects during crisis periods. [Insert Table 6 here] Models (1)-(3), Panel B Table 6 report Fortune firm effects on loan terms for our customer borrowers. Loan amount is significantly larger when the borrower itself is a Fortune firm and when the borrower has larger proportion of purchase from a Fortune firm. The loan spreads is also significantly lower when the borrower itself is a Fortune firm. However, in Models (4)-(5), Panel B Table 6, the analysis on trade credit terms does not have significant Fortune firm effects during crisis periods. Based on the size of the customer-supplier relation, our evidence support that non-fortune firm could benefit from its Fortune partner to obtain better loan terms, such as larger loan amounts and lower loan spread Classifying the quality of firms base on sales revenue Since many high quality firms may not be included in the Fortune 500 list, the way we separate firms into Fortune versus non-fortune may categorize a non-fortune high quality firm into the low quality group. Thus, in this subsection, we follow the Fortune Magazine s logic in creating the Fortune 500 list to use a firm s sales revenue to re-classify firms into high versus low quality. We mark a firm as high quality if its sales revenue is greater than the median level; otherwise, it is marked as low quality. 18

21 [Insert Table 7 here] Table 7 displays the results of our new criterion for high quality firms. Good_partner indicates the borrower itself is a low quality firm with a high quality partner, Good_borrower indicates the borrower itself is a high quality firm with a low quality partner, and Good_both indicates both the borrower and its partner are high quality firms. With this new classification, we do observe evidence consistent with our expectation, but the results become less significant. Such less significant results may indicate that firms benefit from their Fortune partner not only because the Fortune partner has high sales revenue, but also because the Fortune partner offers certification to reveal the firm s quality. 6. Conclusions A firm s supply chain relation may contain information about its business relation, position in the product market, and its future prospects in terms of performance. Although existing literature has identified evidence that such relation may affect a firm s corporate policies and investor wealth, we know little about its impact on a firm s financing activities. In this study, we propose that banks can rely on such relation to identify a borrower s quality and then incorporate it into the decision on the borrower s terms of loans. Based on the idea that high quality firms have stringent requirements in selecting their corporate partners, we believe that a high quality partner in a firm s supply chain helps improve its operations, performance, and its overall quality. Since the Fortune 500 firms are well-known firms and considered as a mark of prestige and high quality, we use them to measure the quality of a firm in the supply chain. Hence, we examine whether a Fortune partner in a firm s supply chain helps 19

22 certify the quality of this firm to its lenders by examining the loan terms. In addition, we also examine whether Fortune firms help their supply chain partners to go through tough times with trade credit arrangements. In support of our expectation, we show that a Fortune partner may help a firm to obtain loans with lower spread and larger amount. We also find evidence that Fortune firms would extend trade credit or speedup payments to help their suppliers/customers during crisis periods. The implication to a non-fortune borrower is that maintaining strong relationship with high quality firms, such as Fortune 500, does help to improve the lender s view on its quality. 20

23 References 1. Bharath, Sreedhar, Sandeep Dahiya, Anthony Saunders, and Anand Srinivasan, 2007, So what do I get? The bank s view of lending relationships, Journal of Financial Economics 85, Bharath, Sreedhar, Sandeep Dahiya, Anthony Saunders, and Anand Srinivasan, 2011, Lending Relationships and Loan Contract Terms, Review of Financial Studies 24, Chava, Sudheer, and Amiyatosh Purnanandam, 2011, The Effect of Banking Crisis on Bank-Dependent Borrowers, Journal of Financial Economics 99, Chava, Sudheer, and Michael Roberts, 2008, How Does Financing Impact Investment? The Role of Debt Covenants, Journal of Finance 63, Cull, Robert, Lixin Colin Xu, and Tian Zhu, 2009, Formal Finance and Trade Credit during China s Transition, Journal of Financial Intermediation 18, Dahiya, Sandeep, Anthony Saunders, and Anand Srinivasan, 2003, Financial Distress and Bank Lending Relationships, Journal of Finance 58, Dahiya, Sandeep, Maniu Puri, and Anthony Saunders, 2003, Bank Borrowers and Loan Sales: New Evidence on the Uniqueness of Bank Loans, Journal of Business 76, Dennis, Steven A., and Donald J. Mullineaux, 2000, Syndicated Loans, Journal of Financial Intermediation 9, Ellis, Jesse A., C. Edward Fee, and Shawn E. Thomas, 2012, Proprietary Costs and the Disclosure of Information about Customers, Journal of Accounting Research 50, Fee, C. Edward, and Shawn Thomas, 2004, Sources of Gains in Horizontal Mergers: Evidence from Customer, Supplier, and Rival Firms, Journal of Financial Economics 74, Files, Rebecca, and Umit G. Gurun, 2014, Lenders Response to Restatements along the Supply Chain, Working paper, University of Texas at Dallas. 12. Fok, Robert C.W., Yuan-Chen Chang, and Wan-Tuz Lee, 2004, Bank Relationships and Their Effects on Firm Performance around the Asian Financial Crisis: Evidence from Taiwan, Financial Management 33,

24 13. Giannetti, Mariassunta, Mike Burkart, and Tore Ellingsen, 2011, What You Sell Is What You Lend? Explaining Trade Credit Contracts, Review of Financial Studies 24, Hertzel, Michael G., Zhi Li, Micah S. Officer, and Kimberly J. Rodgers, 2008, Inter-firm Linkage and The Wealth Effects of Financial Distress Along The Supply Chain, Journal of Financial Economics 83, James, Christopher, 1987, Some Evidence on The Uniqueness of Bank Loans, Journal of Financial Economics 19, Kale, Jayant R., and Husayn Shahrur, 2007, Corporate Capital Structure and The Characteristics of Suppliers and Customers, Journal of Financial Economics 83, Lee, Sang Whi, and Donald J. Mullineaux, 2004, Monitoring, Financial Distress, and the Structure of Commercial Lending Syndicates, Financial Management 33, Love, Inessa, Lorenzo A. Preve, and Virginia Sarria-Allende, 2007, Trade Credit and Bank Credit: Evidence from Recent Financial Crises, Journal of Financial Economics 83, Lummer, Scott L., and John J. McConnell, 1989, Further Evidence on The Bank Lending Process and the Capital-Market Response to Bank Loan Agreements, Journal of Financial Economics 25, Pandit, Shail, Charles E. Wasley, and Tzachi Zach, 2011, Information Externalities along the Supply Chain: The Economic Determinants of Suppliers Stock Price Reaction to Their Customers Earnings Announcements, Contemporary Accounting Research 28, Petersen, Mitchell A., and Raghuram G. Rajan, 1994, The Benefits of Lending Relationships: Evidence from Small Business Data, Journal of Finance 49, Petersen, Mitchell A., and Raghuram G. Rajan, 1997, Trade Credit: Theories and Evidence, Review of Financial Studies 10, Shahrur, Husayn, 2005, Industry structure and horizontal takeovers: Analysis of wealth effects on rivals, suppliers, and corporate customers, Journal of Financial Economics 76, Slovin, Myron B., Marie E. Sushka, and Carl D. Hudson, 1990, External Monitoring and its Effect on Seasoned Common Stock Issues, Journal of Accounting and Economics 12,

25 25. Slovin, Myron B., Marie E. Sushka, and John A. Polonchek, 1993, The Value of Bank Durability: Borrowers as Bank Stakeholders, Journal of Finance 48, Sufi, Amir, 2007, Information Asymmetry and Financing Arrangements: Evidence from Syndicated Loans, Journal of Finance 62, Wang, Jin, 2012, Do Firm s Relationships with Principal Customers/Suppliers Affect Shareholder s income? Journal of Corporate Finance 18,

26 Table 1. The sample distribution by year, Fortune500 Company. Panel A: The distribution of firms in syndicated loan market Supplier Customer Supplier Customer Year # of firms % # of firms % # of deals % # of deals % Total 1, , , ,

27 Table 1. The sample distribution by year, Fortune500 Company.(Conti.) Panel B: The distribution of Fortune 500 Company Supplier Customer Fortune 500 Obs. % Obs. % % % % % Total % % 25

28 Table 2. Variable definitions Category Variable Name Definition TA TA is total assets of firms as reported the current fiscal year prior to the syndicated loan deal active date. Fortune_partner Fortune_partner is a dummy variable that equals to one if firm isn t Fortune500 Company itself but has Fortune500 partner in supply chain. Fortune Fortune_borrower is a dummy variable that equals to one if firm is Fortune500 Fortune_borrower dummy Company itself but doesn t have Fortune500 partner in supply chain. Fortune_both Fortune_both is a dummy variable that equals to one if firm is Fortune500 Company itself and has Fortune500 partner in supply chain. Timing CRISIS is a dummy variable that take the value of one if the syndicated loan CRISIS dummy deal take place in crisis period. CAPEX CAPEX is capital expenditure of firms divided by total assets of firms in the fiscal year prior the loan deal active date. Investment PPE PPE is property, plant and equipment of firms divided by total assets of firms in the fiscal year prior the loan deal active date. CLT CLT is collateral of firms, which equals inventories plus PP&E divided by total assets of firms in the fiscal year prior the loan deal active date. Credit quality RATING_SCORE RATING_SCORE measures the credit quality of firms. The original rating AAA will equal 1, AA+ equals 2 and so on. LVG LVG is total debts of firms divided by total assets of firms in the fiscal year prior the loan deal active date. Leverage LTD LTD is long-term debt of firms divided by total assets of firms in the fiscal year prior the loan deal active date. INTEX INTEX is interest expense of firms divided by total assets of firms in the fiscal year prior the loan deal active date. CASH CASH is cash of firms divided by total assets of firms in the fiscal year prior the loan deal active date. LTD1 LTD1 is long-term debt due in one year of firms divided by total liabilities of firms in the fiscal year prior the loan deal active date. Maturity LTD3 is long-term debt due in three years of firms divided by total liabilities of LTD3 structure firms in the fiscal year prior the loan deal active date. LTD5 LTD5 is long-term debt due in five years of firms divided by total liabilities of firms in the fiscal year prior the loan deal active date. LNS LNS is the natural log of one plus sales of firms as reported the current fiscal year prior to the syndicated loan deal active date. SG is annually sales growth of firms as reported the current fiscal year prior to SG the syndicated loan deal active date. Profitability EBITDA is earnings before interest, tax, depreciation and amortization of firms EBITDA divided by total assets of firms in the fiscal year prior the loan deal active date. TAX TAX is income taxes of firms divided by Pre-tax income of firms in the fiscal year prior the loan deal active date. DYc is dividend yield based on calendar year, which is dividend per share of DYc firms divided by share price of firms as reported the current fiscal year prior to the syndicated loan deal active date. Dividend policy DYf DYf is dividend yield based on fiscal year, which is dividend per share of firms divided by share price of firms as reported the current fiscal year prior to the syndicated loan deal active date. DPO DPO is dividend payout, which is dividend of firms divided by the EBITDA of firms in the fiscal year prior the loan deal active date. TBQ is Tobin s q, which is total assets minus share equity plus market value of Performance TBQ firms then divided by total assets of firms in the fiscal year prior the loan deal active date. 26

29 Table 3. Firm characteristics by borrower type Suppliers Customers Variable Name N Mean Median N Mean Median TA 1,511 2, ,402 36,669 5,033 Fortune Dummy Fortune_partner 1, , Fortune_borrower 1, , Fortune_both 1, , Crisis Dummy CRISIS 1, , Investment CAPEX 1, , PPE 1, , CLT 1, , TBQ 1, , Credit rating RATING_SCORE 1, , Leverage LVG 1, , LTD 1, , INTEX 1, , CASH 1, , Maturity structure of debt LTD1 1, , LTD3 1, , LTD5 1, , Profitability LNS 1, , SG 1, , EBITDA 1, , TAX 1, , Dividend policy DYc 1, , DYf 1, , DPO 1, , Variable are defined as: TA is total assets of firms as reported the current fiscal year prior to the syndicated loan deal active date. 27

30 Fortune_partner equals to one if firm isn t Fortune500 Company itself but has Fortune500 partner in supply chain. Fortune_borrower equals to one if firm is Fortune500 Company itself but doesn t have Fortune500 partner in supply chain. Fortune_both equals to one if firm is Fortune500 Company itself and has Fortune500 partner in supply chain. CRISIS equals one if the syndicated loan deal take place in crisis period. CAPEX is capital expenditure of firms divided by total assets of firms in the fiscal year prior the loan deal active date. PPE is property, plant and equipment of firms divided by total assets of firms in the fiscal year prior the loan deal active date. CLT is collateral of firms, which equals inventories plus PP&E divided by total assets of firms in the fiscal year prior the loan deal active date. RATING_SCORE measures the credit quality of firms. The original rating AAA will equal 1, AA+ equals 2 and so on. LVG is total debts of firms divided by total assets of firms in the fiscal year prior the loan deal active date. LTD is long-term debt of firms divided by total assets of firms in the fiscal year prior the loan deal active date. INTEX is interest expense of firms divided by total assets of firms in the fiscal year prior the loan deal active date. CASH is cash of firms divided by total assets of firms in the fiscal year prior the loan deal active date. LTD1 is long-term debt due in one year of firms divided by total liabilities of firms in the fiscal year prior the loan deal active date. LTD3 is long-term debt due in three years of firms divided by total liabilities of firms in the fiscal year prior the loan deal active date. LTD5 is long-term debt due in five years of firms divided by total liabilities of firms in the fiscal year prior the loan deal active date. LNS is the natural log of one plus sales of firms as reported the current fiscal year prior to the syndicated loan deal active date. SG is annually sales growth of firms as reported the current fiscal year prior to the syndicated loan deal active date. EBITDA is earnings before interest, tax, depreciation and amortization of firms divided by total assets of firms in the fiscal year prior the loan deal active date. TAX is income taxes of firms divided by Pre-tax income of firms in the fiscal year prior the loan deal active date. DYc is dividend yield based on calendar year, which is dividend per share of firms divided by share price of firms as reported the current fiscal year prior to the syndicated loan deal active date. DYf is dividend yield based on fiscal year, which is dividend per share of firms divided by share price of firms as reported the current fiscal year prior to the syndicated loan deal active date. DPO is dividend payout, which is dividend of firms divided by the EBITDA of firms in the fiscal year prior the loan deal active date. TBQ is Tobin s q, which is total assets minus share equity plus market value of firms then divided by total assets of firms in the fiscal year prior the loan deal active date. 28