Institution: University of Amsterdam, Amsterdam Business School
Program: MSc Business Economics, Finance
Document type: Master Thesis
Title: Alternative debt financing during a crisis: A study on trade credit of firms in Asia
Name of author: Amira Dianti
Date: July 2014
Alternative debt financing during a crisis: A study on
trade credit of firms in Asia
By Amira Dianti Amsterdam Business School
Abstract
The event of a financial crisis allows us to study financing behaviors by firms during changing economic conditions. One of the major concerns in debt financing during a financial crisis is a firm’s capital access to traditional bank lending causing an increase in demand for alternative external financing. Trade credit is one of the most popular funding for firms that have limited access to the financial capital markets. This study focuses on the use of trade credit and its relationship to firm size, firm growth and market-wide factors. I found that large firms that enjoy better access to the capital market not only increase accounts payables during a crisis, but also accounts receivables. Trade credit is also an important channel to manage firm sales growth, especially for small firms that are commonly more constrained to alternative financing sources. Lastly, this study looks into the dependence of trade credit on bank sector development especially for emerging countries with less developed financial markets. The findings highlight the importance of access to bank credit especially in times of financial crisis.
Content
1. INTRODUCTION & BACKGROUND 3
1.1 The Asian financial crisis 5
2. LITERATURE REVIEW 5
2.1 Bank loans during the crisis 6
2.2 Trade credit during the crisis 9
2.3 Trade credit and firm size 11
2.4 Trade credit and firm growth 11
2.5 Trade credit and economy development 12
3. DATA & METHODOLOGY 14
3.1 Sample 14
3.2 Empirical approach 15
4. DESCRIPTIVE STATISTICS 17
5. RESULTS 19
5.1 Trade credit and firm size 19
5.1 Trade credit and firm growth 21
5.3 Trade credit and economy development 23
1. INTORDUCTION & BACKGROUND
A firm’s source of capital is dependent on the willingness of funders to lend and the demand of investment finance of the firm (Bhaird, 2012). The have been several studies that had done research on the changes in capital structure of firms after a financial crisis1. De Meza and Webb (1987) suggested that debt is the equilibrium source of finance during an economy downturn as there is investment in excess of socially efficient levels due to over-lending. One of the major concerns in debt financing during a financial crisis is a firm’s capital access to traditional bank lending. The dependence a firm has on bank financing puts them in a vulnerable position when an unforeseeable event disrupts the financial system and resulting in a credit crunch. During a crisis, firms become more concerned about debt exposure and lenders become stricter in giving credit. As a result of this bank lending constraint, firms will have higher demand to alternative financing options. With most financial systems being bank-centric, there is great need to have alternative funding sources especially during a financial crisis. This highlights the importance of studying alternative sources of funds for firms experiencing a financial crisis.
Astrauskaite & Paskevicius (2014) studied the relationship between banks and debt markets and found that rises in bank lending restrictions caused by a financial crisis have increased the use of non-bank credit. They further explain that during a crisis, investors invest out of fear of losing money rather than profit seeking and would typically search for yield in low interest rate environments. However, not all markets have the ease of having such diversified debt markets. For example, in developed countries where the bond market is more mature, firms are more exposed to alternative sources of funds. Thus, the use of bank loans might not be as large as those in emerging countries. Different types of firms will also have different access to alternative capital markets.
Prior studies have found that trade credit is a popular source of alternative financing where there is a bank lending constraint2. Casey and O’Toole (2014) found that firms that are more restricted in access to credit are more likely to use trade credit rather than other financing alternatives. In other words, when firms are experiencing a credit crunch and it
1 See, for example, Deesomsak et al. (2009)
has limited sources of alternative financing, it will rely of trade credit as a source of fund. For firms that enjoy better access to the capital market, however, taking on credit could be an unattractive financing option because of its high costs. But it also means it makes it attractive for them to extend credit to customers that are constrained. In this situation, trade credit then becomes a driver of firm performance when these firms are experiencing a lack of funding source as trade credit can encourage sales of goods that otherwise would not have been made.
The use of trade credit during a financial crisis is stronger in emerging countries for two reasons (Daniel and Scott, 2004). Firstly, trade credit is a major portion of liabilities for firms in emerging countries, suggesting that it is an important source of finance in these firms. Secondly, firms in emerging countries show vulnerability to external risks. Furthermore, the use of trade credit is also highly dependent on the access to funds that can finance credit extended to customers. Therefore, it is expected that the use of trade credit will depend on market-wide factor that influence the supply of credit for these firms.
The aim of this study is to examine how firm in Asian countries affected by the Asian crisis in 1997 change their borrowing behavior during the crisis. The types of debt I will focus on are bank debt and trade credit. Furthermore, I will also study how a firm’s use of trade credit can affect its performance. With this research I will answer and discuss the four questions below:
1. Do large firms have higher net trade credit after a crisis by reducing credit from suppliers and extending credit to customers?
2. Do small firms have a higher dependence on net trade credit to manage growth after a crisis?
3. Is the sensitivity of banking sector development to net trade credit larger in emerging countries?
I will take samples of the three firms most affected by the crisis, which are Thailand, Indonesia and South Korea. For comparison, I will also take samples of countries that were least affect by the crisis, which are Australia, Singapore and Japan. I will use both firm-specific and market-wide (such as economy conditions) determinants to have a better
insight into the debt choices of these firms. To my knowledge, there is very little literature that studies trade credit and firm performance with countries affected by the Asian crisis. By having a deeper insight into changes in financing behavior after a crisis, companies can better identify financial risks and manage it in changing economic conditions.
1.1 The Asian financial crisis
The Asian financial crisis in 1997, also known as the “Asian Contagion”, had one of the most detrimental effects to the economies of Asian countries. Asia’s growing economy has become a driving powerhouse to the world’s economy. Its fast growing trend has made them an attractive target to both domestic and foreign investors. Starting from the early 1990’s, many domestic banks in these underdeveloped economies found it attractive to borrow money abroad to finance domestic investments3. Over time, these national banks found themselves borrowing excessively from abroad and at the same time lending excessively at home.
When the Thai government was forced to float the Thai Baht due to the collapse of the currency and lack of foreign exchange, Thailand had no choice but to post for a multi-billion dollar bailout from the IMF. The crisis then spread to other countries in Asia, but the other two countries that were most affected by this crisis was South Korea and Indonesia. The Asian financial market suffered severe devaluation in stock and asset prices as both domestic and foreign investors started fleeing the country. As a result, lenders became alarmed and started withdrawing credit. With a sudden switch in market confidence, it made it difficult for the financial market to allocate funds efficiently. The foreign loans that were financing domestic bank lending were no longer available. National banks then found themselves with abundance of domestic loans and a lack of funds to finance them. Furthermore, firms affected by the crisis will delay investments and start delevering which further contracts credit flow in the market. This case was most severe in countries that had less mature economies. With the change in both supply and demand of capital and access to financial markets caused by the crisis, it can be expected that the financial crisis have great impact in changing patterns of debt structure in firms
3 See Goldstein (1998)
1. LITERATURE REVIEW
2.1 Bank loans during a crisis
Many prior studies have found that there is a decrease in use of bank lending after a financial crisis. There have also been many theories formed to explain this decrease in a firm’s use of bank debt. A study by Bhaird (2012) who studied the demand and supply of debt and equity after a financial crisis suggests that prudent lending is the cause of the decrease in overall bank lending. When foreign investments that were funding these domestic loans were no longer available, banks were left with increasing domestic loans with a lack of funding to finance them. Banks, then, become stricter in giving loans and decreases lending to firms with poor credit quality which commonly increases during an economy downturn. Also, banks have to retain capital and strengthen capital adequacy ratios due to greater capital reserve requirements and a need to shrink balance sheet (Honohan, 2008). Moreover, firms become exposed to adverse selection problems and these impacts are more severe in developing countries due to a lack of diversification of sources of external funding (Vos et al., 2007). The impact of a financial crisis towards bank lending activities allows us to study the borrowing behaviors of firm experiencing the crisis. Before we discuss alternative sources of fund, let us first look into how firms borrowing behavior to banks change during a crisis.
Table 1 summarizes the relationship between bank loans and different firm-specific determinants. There have been various results to studies of specific firm characteristics to observe financing patterns of firms. Prior studies suggest that firm characteristic determinants are more important because of pro-cyclical lending behavior of financial institutions4.
4 See, for example, Deesomsak (2009)
Table 1
Bank loans and firm-specific determinants
The dependent variable is Bank loans consist of (i) revolving bank debt, which includes committed revolving credit facilities and (ii) term bank debt, which includes term loans and other informal credit scaled by total liabilities of firms listed in Datastream. Bank loans are scales by total liabilities. MTB is market-to-book which is market capitalization over total market-to-book value. SIZE is constructed by taking log of sales. PROF is profitability measured by EBITDA over total assets. Proxy for TANG (tangibility) is measured by total tangible assets over total assets. The standard errors are obtained using clustering on the country and time as explained in the paper. Absolute values of robust t-stats are in brackets.
Most affected Least affected
Prior During Post Prior During Post
MTB -0.0239419 0.1193114 -0.1613934 0.0568932 0.0136087 -0.0028594 (-0.44) (1.14) (-4.79) (1.06) (0.17) (-0.08) SIZE -0.0246967 -0.0153392 -0.0459175 -0.017382 -0.09058959 -0.0256341 (-6.08) (-2.54) (-3.88) (-4.75) (-0.99) (-0.89) PROF 0.2333525 -0.8552278 0.19674 0.0941221 -0.2893195 0.113469 (1.97) (-2.12) (1.34) (0.95) (-0.81) (0.86) TANG 0.0012827 0.0221166 0.0694486 -0.018063 -0.0327979 0.016282 (0.04) (0.36) (2.54) (-0.64) (-0.57) (0.64) Cons 0.3685897 0.3314453 0.6971007 0.1884684 0.4087599 0.3096016 (5.99) (2.62) (13.78) (2.36) (3.51) (5.98)
As shown on the results, firm size and profitability are two of the firm-specific determinants that are statistically significant during the crisis period for firms in countries that were most affected by the crisis. Looking at profitability, its positive correlation with bank loans increased in number and significance when the crisis hits. The signaling
hypothesis suggests that when a firm is profitable and able to build reputation of high
earning, firms are less constrained to access of financial markets. One of the aftermaths of a financial crisis is the tightening of bank regulations in order to recover financial stability. As a result, banks are forced to be stricter in granting loans and they will start rationing creditors. This means, only firms with the highest credit quality applying for loans will be granted, while poor credit quality firms will be rationed. Carey et al. (1998), who studied the lending behavior of corporate and financial institutions, suggested that, in a crisis, bank regulators impose high loan loss reserve requirements to banks for loans to firms with low credit quality. As a result, banks are discouraged to grant loans to firms with low credit quality. With little access to bank loans, firms with low credit rating will choose to issue non-bank debt while high credit rating firms enjoy more access to bank loans.
For firm size on the other hand, the correlation with bank loans are negative and statistically significant for all three periods. This is contradicting to a study by Lin (2007) who suggested that 98% of SME firm were unable to have access to bank loans in 2006. One reason that could explain this behavior is relationship-based lending that banks have with small firms. Relationships between banks and small firms tend to be a lot closer than their relationships with larger banks. Cole, Goldberg, and White (1999) studied lending behavior of large banks to small business and found that large banks tend to favor transactions-based lending when dealing with small firms while smaller banks use relationship-based lending. This transaction-based lending usually rely on hard information obtained by monitoring technologies. While small banks tend to rely on soft information. Given the existence on smaller banks are greater in developing countries, we can expect this relationship-based lending to small firms to be common in these countries. Moreover, large firms are commonly more able to commit to long-term debt due to easier access to the capital market, while smaller firms depend majorly on bank debt. Also, asymmetric information problems are commonly greater for small firms that large firms, thus, making it harder to small firm to have access non-bank loans. The impact slightly increases during the crisis period, which is in line with the findings of Berger, Klapper,
and Udell (2001) that suggested that when banks are experiencing financial distress, they are less likely to lend to small firms due to increasing asymmetry information.
With the tightening of bank regulations during a crisis, we can expect firms to move to alternative sources of debt. When liquidity is unrestricted, firms prefer to take on cheaper bank loans to more expensive loans from non-institutional sources. However, when liquidity dries up and firms are faced with bank lending constraints, they are more willing to pay a premium on these non-institutional loans (Casey and O’Toole, 2014). In this study, I will focus on trade credit as the most common alternative financing for firms during times of contractions. In the next section, I will discuss more extensively about the use of trade credit by firms experiencing a crisis. We have also seen how changes in bank lending behavior changes during a crisis and this is expected to affect the use of alternative financing of firms. Therefore, I will also look into how trade credit relates to firm size, firm growth and its dependence on market-wide factors (such as banking sector).
2.2 Trade credit during the crisis
Trade credit is defined as a short-term loan a supplier gives to its customers for a purchase of its products. It is expressed in the balance sheet as a percentage of accounts payable to total liabilities. There are a couple of motives for companies to take on trade credit. Nilsen (1999) suggested that trade credit can act as a transaction service allowing higher cost savings when customers purchase more of the supplier’s products. Moreover, Schwartz (1974) also found that it is attractive for customers to take on trade credit because it simplifies cash management to its firm. From a supplier’s point of view, trade credit allows firms to reduce idle cash holdings and hold interest-earning assets instead (Ferris, 1981). With that being said, we can easily see why trade credit would be a popular source of debt for companies.
Nevertheless, there are disadvantages to trade credit. Nilsen (1999) mentioned in his study that, during normal economy conditions, trade credit is an unattractive substitution for bank loans. Firstly, the terms of trade credit are more restricted as it is associated with purchased goods while bank loans are more unrestricted. Secondly, repayments of trade credit are usually short-term while bank loans are more flexible with repayment terms.
Lastly, because credit provider is a supplier rather that an independent financial institution, customers face the risk of high late payment penalties. Furthermore, customers can risk damaging the long-term relationship it has with its suppliers, which could be very costly to a firm’s operations. In other words, when a firm’s source of debt is unrestricted, trade credit becomes a rigid and costly choice of debt.
Although trade credit maybe unattractive for firms during normal economy conditions, it becomes significantly demanded during a credit crunch. While other sources of debt become more scarce and rigid during a financial crisis, trade credit is readily available at hand. Blazenko & Vandezande (2003) suggested that a common idea to use trade credit is to increase sales. Furthermore, suppliers can use trade credit to encourage economies of scale, i.e. encourage customers to purchase in large amounts. This is highly valued especially for firms that have limited access to alternative sources of funds. Burkant and Ellingsen (2014) also found that in situations where banks are rationing credit, suppliers are the best source of credit as they have relative advantages to overcome firm related moral hazard and information asymmetries. Moreover, Nilsen also explains that customers are willing to pay a premium on interest rate for the used of short-term funds. Therefore, we can expect an increase of trade credit during a financial crisis particularly for firms that are restricted to alternative sources of financing.
Love et al. (2007) explained that net trade credit measure the willingness of firms to extend trade credit to customers s firms that obtain more trade credit from suppliers are more willing to extend credit to customers. To examine this, I will use three measurements of trade credit mentioned in Love et al. (2007) study. Accounts payable and accounts receivable are the two main variable of interest, which show the amount of trade credit that firms obtain from its suppliers and extend to its customers, respectively. The difference between the two variables measures the willingness of firms to provide credit to customers net of credit from suppliers.
Trade credit is commonly used as a tool to promote sales to customers during times of contractions. In the next section I will discuss further on how firms use trade credit to manage growth of sales.
2.3 Trade credit and firm size
Wilner (2000) found an increase in the use of trade credit in times when credit rationing increases. However, this might not be true for all firms. I have discussed earlier how trade credit can be costly to a firm. Better access to the capital market allows large firms to be less dependent on costly trade credit from suppliers and is more physically able to extend credit to their customers. Therefore, I hypothesized that, after the crisis, large firms that have better access to bank loans will use less trade credit even after the crisis. With the high cost associated with trade credit, we can expect a reduction of trade payables for profitable firms. However, at the same time, I expect these firms to have an increase in trade receivables, as they are more willing to extend credit to their customers. Nilsen (2002) explains that firms that have better access to the credit market are in the best position to financially support suppliers and customers through extension of credit. Moreover, there have been proof that firms who have a higher liquidity cushion (cash and cash flow generation assets) are better able to support the commercial operations of their customers and have less dependence on credit from suppliers (Love et al., 2007). On the contrary, the opposite is expected for smaller firms. With limited access to bank loans, smaller firms are more reliable on credit from suppliers as it is easier to access and more available at hand. With that being said, I expect an increase in trade payables for smaller firms. At the same time, however, firms will delay bill repayments and lengthen credit durations to suppliers, hence, increasing trade payables Love et al. (2007).
2.4 Trade credit and firm growth
There have been many studies relating trade credit with firm performance. A study by Ferrando and Mulier (2013) suggested that firms use both accounts receivables and accounts payables to support firm performance. Where accounts payables are able to lessen dependence on financial market imperfections, accounts receivables are able to do so in the product market. They also found that firms that are more vulnerable to financial market shocks and market imperfections depend more on trade credit to manage firm performance. They base their study on firms in the euro area and found that trade credit has been a stable source of financing for these firms, however, the use of it tends to
decline when bank credit becomes more accessible. With this, it would be interesting to see how firms use trade credit to manage firm performance during a credit crunch.
One common indication of firm performance is firm growth. Many researches have found that growth of firms is negatively correlated with size of firm as smaller firms have more room to grow than larger firms5. The ability of a firm to grow also relies on the ability of these firms to obtain capital in order to expand and grow their businesses. In countries were the financial markets are more developed, smaller firms are able to grow faster than those with less developed financial markets. This is because asymmetric information is a common problem for small firms seeking external financing to fund potentially profitable growth. Moreover, in countries that have more mature credit markets, financial contractions are more relaxed, so these small firms are better facilitated to grow. This relationship between firm growth and the capital market gives interesting background to study the use of trade credit to manage firm growth when the financial market experiences a contraction.
Both large firms and small firms use a wide amount of trade credit. Prior research has found that large firms use trade credit as a “cash management control” by delaying payments to suppliers and pushing cash collection from customers (Ferrando and Mulier, 2013). For small firms, however, the use of net trade credit during times of scarce funding becomes of great importance to its daily operations. It has been discussed in previous sections how trade credit can serve as a marketing device to promote sales to customers, especially when credit is limited. These firms use accounts payables to finance its production and accounts receivables to allow sales of its products to constrained customers. Therefore, it can be expected, that in emerging countries where the financial markets are less developed, small firms are highly dependent on trade credit to manage its firm growth.
2.5 Trade credit and economy development
I have discussed how the use of trade credit can be different for different types of firms using firm-specific determinants. However, a firm’s choice of debt also depends strongly
on country specific determinants such as economic conditions. Changing economic conditions will affect a firm’s preference of capital source (Fan et al., 2004). In this study, I will use market-wide determinants use in Deesomsak et at. (2008), who studied the determinants of debt maturity after a financial crisis.
Table 2
Market-wide determinants Measurement
Economy development (EDEV)
Dummy equal to 1 for developing economy and 0 otherwise
Bank development (BKDEV) Bank assets/GDP Stock market development (MKDEV)
Market capitalization/GDP Government bond Term structure of interest rates (TERM) Yield − treasury bill (3 month rate)
Inflation (INF) Changes in consumer price index
One of the most significant determinants that were found in Deesomsak’s study was economy development. The financial crisis will impact different countries in different ways depending on the economy development level of the country. A study by Essers (2013) suggested that the impacts of financial shocks are most detrimental to developing economies through inflationary pressure and major reduction in growth. For developed countries, financial shocks are commonly short-lived and relatively easier to recover from. With that being said, it is expected that the impact of the financial crisis will also change the borrowing behavior of firms in developed and emerging countries in different ways.
In emerging countries where the financial markets are less mature firms commonly resort to bank credit as their primary source of fund. Therefore, when the financial system is disrupted, the impact on changes in firm borrowing behavior will be larger in these emerging countries. In countries where the banking sector is less developed, banks are more constrained in lending money to creditors during a crisis. As a result, firms in these countries will rely more on informal non-bank credit (i.e. trade credit) in a crisis (Li et al., 2013). There have been several opposing finding towards the use of trade credit for firms experiencing a financial crisis. Love et al. (2007) found that firms in economies that are more financially vulnerable to the crisis are more likely to cut the supply of credit to customers and increase the use of credit from suppliers. In countries where the financial
stability was severely hit and the credit crisis was most prominent, the banking system will experience a panic as banks start to withdraw credit to borrowers. As a result, over time, panicked supplier will shrink investments in their customer relationships by withdrawing trade credit and tightening credit terms (Friedman and Schwartz, 1963). On the contrary, many researchers have found that firms in developing countries depend on trade credit to fund its production during a crisis. Cunat (2001) suggested that fast growing firms use trade credit to finance its operations when other sources are not sufficiently available. Furthermore, firms that are experiencing liquidity shortage will try to overcome it by passing on the shock to their suppliers (Boissay and Gropp, 2007). Unlike firms in developed countries, where the financial markets are more mature, firms in emerging countries have more limited alternative funding source. Therefore, its reliance on the use of trade credit will become higher in a crisis. In any case, we can see that the use of trade credit depends on the development of its financial markets.
The supply side of the theory also suggest that trade credit will also depend on the access to fund to finance the credit. In emerging countries where the financial markets are less developed, bank loans will be the primary source of external financing. Therefore, firm’s ability to extend credit to customers will depend on their access to bank loans in these countries. With that, I argue that the sensitivity of bank development to net trade credit is larger in emerging countries during a crisis. In emerging countries where banking sector is less developed, small firms that have limited access to alternative sources of finance will use more trade credit as it is the most accessible and readily available source of finance for these firms.
2. DATA & METHODOLOGY
2.5 Sample
This study will be based on a major currency crisis that occurred in Asia during 1997, also known as the “Asian Contagion”. The sample includes publicly traded firms in countries most affected by the crisis (Thailand, Indonesia and South Korea). To control for differences in the impact of the crisis, this study will also include countries least
affected by the crisis (Australia, Singapore and Japan). Balance sheet information is obtained from COMPUSTAT. All financial institutions are excluded from the sample. This study will consist of three periods, prior, during and post financial crisis. Prior period covers the years 1994-1996 and during crisis period is 1997. The after crisis period covers the years 1998-2000. Studies have found that firms in countries that were affected by the Asian financial crisis were fully recovered after the third year after the crisis6. Therefore, the after crisis period captures the recovery period of these countries. This study uses 1,870 observations from 6 countries.
2.6 Empirical approach
The first part of this research looks into the use of trade credit and firm size after the crisis. It is expected that large firms with better access to the capital market are more willing to extend trade credit to its customer (increasing accounts receivable) and will reduce costly credit from suppliers (decreasing accounts payable). In this study I will use three measures of trade credit7.
1. Trade Receivable (TR) = Trade Receivables / Total Sales (1) 2. Trade Payable (TP) = Trade Payables / Cost of Goods Sold (2) 3. Net Trade Credit (NTC) = (TRR – TPR) / Total Sales (3)
Trade credits are scaled using Total Sales (for receivables) and Cost of Goods Sold (for payables). These ratios show the importance of trade credit in the financing activities of a firm. Scaling by flow variables controls for changes in economic activities (i.e. reduction of sales that are commonly associated with a crisis). TR measures the percentage of goods sold on credit while TP measures the percentage of products purchased on credit. NTC measures the willingness of firms to extend credit to customers net of its credit from supplier. Firms that receive more credit from suppliers are more willing to extend credit to customers. Below is the regression equation used to test this hypothesis:
𝑌!,! = 𝛽!+ 𝛽!𝑃𝑅𝑂𝐹!!!+ 𝛽!𝐿𝐴𝑅𝐺𝐸!!!+ 𝛽!𝑀𝑇𝐵!!!+ 𝛽!𝑇𝐴𝑁𝐺!!!+ 𝜀 (4)
6 See, reference, Love et al. (2007)
Yi,t is the dependent variable, which is one of the three trade credit measurement mentioned in equation x. A dummy variable is constructed for large firms, which are characterized by large amounts of log (sales). The dummy (LARGE) is classified based on their log (sales) falling above the 75th percentile. Firm specific determinants mentioned on Table 2 are included in the regression to control for firm heterogeneity. The error term 𝜇i,t is time varying and serially uncorrelated with mean zero and variance 𝜎2. The explanatory variables are lagged one period to control for possible endogeniety problems. To ensure the robustness of the results, I restrict the data of the key variables and remove outliers. Firstly, I remove all observations that seem to be misreported (i.e. negative trade payables). Secondly, trade credit ratios that imply trade credit is longer than one year are also eliminated.
The second part of this study looks into the relationship between the use of trade credit for small firms and the firm’s growth. Previous literature has found that trade credit can serve as a marketing device to encourage sales to constrained customers. Small firms, which are characterized by a small amount of log (sales), tend to have more constraints in obtaining alternative source of finance, thus, having a higher reliance on trade credit to make sales. At the same time, cash management theory suggests that large firm use a large amount of both accounts payable and receivable. To test the behavior of firms with different sizes, I construct a dummy variable (SMALL) where firms are classified based on their log (sales) falling below the 25th percentile.
𝐺𝑅𝑂𝑊𝑇𝐻!,! = 𝛽!+ 𝛽!𝐺𝑅𝑂𝑊𝑇𝐻!!!+ 𝛽!𝐵𝐾𝐿𝑂𝐴𝑁!+ 𝛽!𝑆𝑀𝐴𝐿𝐿!+ 𝛽!𝑁𝑇𝐶! + 𝛽!𝑁𝑇𝐶𝑆𝑀𝐴𝐿𝐿!+ 𝜀
(5) The dependent variable, GROWTH, is measured as the annual growth in profit (loss). Coluzzi et al. (2012) found that access to bank loans is an important determinant of firm growth. In this study I add BKLOAN in the regression, which is measured as total short-term and long-short-term loans scaled by total assets. NTC is constructed by the sum of accounts receivable and accounts payable, scaled by total sales. To test the dependence of net trade credit to firm growth for small firms, I include an interaction variable, which is constructed by multiplying NTC and SMALL which is the firm size dummy. A
significant coefficient of this variable indicates that small firms have high dependence towards NTC to manage its sales growth. The error term 𝜇i,t is time varying and serially uncorrelated with mean zero and variance 𝜎2.
The last part of this research examines the sensitivity of banking sector development and the use of net trade credit for firms in emerging countries. A decrease in bank loan access will lead to a decrease in trade credit due to lack of credit supply8. Therefore it is expected that firms in emerging countries, where financial markets are less developed, firms will highly depend on bank sector development to extend trade credit. The dependent variable is net trade credit of firm i at time t, which is constructed using the measurement mentioned in Eq. 3. Table 2 summarizes the market-wide determinants along with the measurement that are included in this regression. To take into account differences in economy conditions, I construct a dummy variable for economy development (EDEV), which equals to 1 for emerging countries (Thailand, Indonesia and South Korea) and 0 for developed countries (Australia, Singapore and Japan)9. An
interaction variable is included in the regression to capture the sensitivity of banking sector development towards net trade credit for emerging countries. A positive and significant coefficient on EBKDEV indicates a strong dependence of trade credit on bank sector development for firms in emerging country. Below is the regression equation to test this hypothesis:
𝑁𝑇𝐶!,! = 𝛽!+ 𝛽!𝐸𝐷𝐸𝑉!!!+ 𝛽!𝑀𝐾𝐷𝐸𝑉!!!+ 𝛽!𝐵𝐾𝐷𝐸𝑉!!!+ 𝛽!𝐼𝑁𝐹!!! + 𝛽!𝐸𝐵𝐾𝐷𝐸𝑉!!!+ 𝜀
(6)
4. DESCRIPTIVE STATISTICS
In general, firms will delever their balance sheet when experiencing a financial crisis. With most firms being financially dependent on bank loans, changes in debt structure are expected when the financial system experiences a crunch. Table 3 Panel B summarizes the average amount of bank loans of firms in the sample countries before, during and after the crisis. As shown on the table, there was a sharp decrease in bank loans, from U$175m to U$76m, during the crisis is countries that were most affected by it. In the period after
8 See Love et al. (2007) 9 See Deesomsak et al. (2009)
the crisis, bank loans increased slightly, however, it was not at the level it was before the crisis. This highlights the importance of studying the changes in debt structure of firms that experience a financial crisis and alternative sources of finance.
In Table 3 Panel B, presents the mean of trade credit ratios before, during and after the crisis. For all three ratios, there was a similar reaction when the crisis hit, namely, a sharp increase in the crisis year. Trade receivables and net trade credit both decreased slightly after the crisis, however, the ratio was still higher than it was before the crisis. For trade payables, however, the ratio increased slightly, even in the after crisis period. This is inline with the theory that firms experiencing bank-lending constraints and have limited access to alternative financing, will rely on trade credit from suppliers. In the next section, I will present a more extensive discussion of the results of the empirical approach used for this study.
Table 3
Panel A Number of observations by country
This table presents the number of observations by country, based on the number of non-missing values of the variable
Countries Number of observations
Most affected Indonesia 133 Thailand 141 South Korea 324 Least affected Australia 248 Japan 942 Singapore 82
Panel B Summary statistics
The table below presents the mean of each variable in the three periods of this study. Bank loans consist of (i) revolving bank debt, which includes committed revolving credit facilities and (ii) term bank debt, which includes term loans and other informal credit scaled by total liabilities of firms listed in Datastream. Bank
debt is the proportion of bank loans to total liabilities. Trade receivables is measured as trade receivables/net
sales, Trade payables is trade payables/cost of goods sold, and Net trade credit is (i.e., receivables minus payables)/net sales. The sample is the unbalanced panel of firms. Prior period covers 3 year before the crisis, during period cover 1997 and post period covers 3 years after the crisis.
Most affected Least affected
Prior During Post Prior During Post
Bank loans 175.11 76.81 92.8 186.9 283.6 285.3 Bank debt 0.24998 0.44016 0.27812 0.414 0.078 0.59 Accounts receivables 464,711 630,020 539,098 42,927 8,484 171,996 Accounts payables 213,887 310,471 396,784 36,532 10,761 120,514 Trade receivables 0.158 0.172 0.143 0.126 0.150 0.192 Trade payables 0.118 0.186 0.257 0.129 0.117 0.208
Net trade credit -1.860 -4.980 -0.080 2.190 -1.830 -6.993
5. RESULTS
5.1 Trade credit and size of firm
Table 4 represents the main results of the first regression equation. I study the coefficient of the large firm size dummy that shows the relationship between trade credits and large firms. The results on average are in line with the theories formed in the previous sections. There was a major reduction of NTC of large firm during the crisis period for firms in countries that were most affected by the crisis. The coefficient of large firms on NTC become negative when the crisis hit and returned positive in the after crisis period. However, I did not find any significance for either of the coefficients. A simple explanation for the sharp increase in the amount of trade credit by large companies during the crisis could be delayed repayments by customers due to temporary illiquidity and
cause credit to accumulate over time. Credit will continue to accumulate until suppliers write off the debt or customers resume to make repayments.
Table 4
Trade credit and size of firm
The dependent variables are the trade credit measures (Eq. 1,2,3 for definition). Size dummy is constructed for large firms by taking the 75th percentile of log of sales. The models are estimated with firm-fixed effects
(see table 2 in the paper) using an unbalanced sample. The standard errors are obtained using clustering on the country and time as explained in the paper. Absolute values of robust t-stats are in brackets.
Most affected Least affected
Prior During Post Prior During Post
Net trade credit
Size dummy 0.000002 -0.00001 0.00003 0.00003 0.000000 0.00002 (-0.95) (-1.44) (1.81) (0.78) (1.21) (1.78) Market to book 0.00001 -0.00002 0.00005 0.00002 -0.00002 0.00003 (0.81) (-1.09) (0.67) (0.83) (-1.03) (0.51) Profitability -0.00002 -0.00002 0.00025 -0.00002 -0.00004 0.00028 (-0.74) (-0.46) (1.16) (-0.81) (-0.65) (1.40) Tangibility -0.00001 0.00006 -0.00009 -0.00001 -0.00008 -0.00009 (-1.28) (0.09) (-2.09) (-1.30) (0.14) (-2.00) Constant -0.00001 0.00002 -0.00006 -0.00001 0.00002 -0.00006 (-0.44) (1.20) (-1.01) (-0.49) (1.12) (-0.96) Trade receivables Size dummy -0.04123 0.03601 -0.03653 0.02358 -0.00808 -0.01599 (-3.51) (2.27) (-2.91) (0.40) (-0.24) (-2.19) Market to book -0.0097 -0.09159 0.02008 -0.00963 -0.09319 0.03143 (-0.24) (-1.61) (0.70) (-0.24) (-1.60) (1.11) Profitability 0.28904 0.39733 -0.00417 0.24254 0.4456 -0.03065 (3.09) (2.20) (-0.09) (2.64) (2.50) (-0.67) Tangibility -0.04411 -0.0038 -0.06491 -0.04581 -0.00751 -0.06905 (-1.69) (-0.11) (-4.53) (-1.71) (-0.22) (-4.84) Constant 0.16325 0.21186 0.19509 0.1581 0.21739 0.19073 (4.30) (4.14) (7.07) (4.18) (4.11) (6.97) Trade payables Size dummy -0.00586 0.20035 0.17802 -0.02778 0.000109 -0.03006 (-0.30) (2.31) (2.52) (-0.49) (-1.67) (-2.76) Market to book -0.23393 0.00024 0.06333 -0.23306 -0.01888 0.00733 (-2.02) (0.70) (1.06) (-2.01) (-0.09) (0.12) Profitability 0.37325 1.3001 0.0221 0.36741 1.5781 0.17608
For firms in countries that were most affected by the crisis, the size coefficient becomes positively and statistically significant against trade receivables when the crisis hits. However, this effect only lasted during the crisis period and sharply decreases after the crisis period. This finding is in line with the theory mentioned in the literature review that large firms that enjoy better access to the capital market are more willing to extend credit to their customers. At the same time, the relationship between trade payables and large firm becomes positive and significant when the crisis hits and stays positive even in the after crisis period. The coefficient slightly decreased in the after crisis period but the relationship was still positive. This finding contradicts the theory that was mentioned in the previous section. With the risk of delayed repayments on trade credit and other reason mentioned in section, the cost of taking on credit from suppliers becomes costly during a crisis, thus, making it an unattractive financing option for large firms. One suggestion that could explain this finding is the bargaining power large firms have over smaller firms, which allows them to force smaller firms that are credit constrained to extend credit (Murfin and Njoroge, 2013). The dominating power that larger firms have on small firms
5.2 Trade credit and firm growth
Table 5 summarizes the main results of the second regression. To see the use of trade credit of small firms to manage growth, we examine the coefficient of the interaction variable between net trade credit and the small firm size dummy variable. The estimations confirm that in countries most affected by the crisis, trade credit is positively and significantly correlated to firm’s growth in general, however, the impact is greater for small firms. The coefficient on the interaction variable becomes statistically significant during the crisis and after the crisis period. The coefficient of the interaction variable decreases in the after crisis period but remained statistically significant at the 5% confidence level. I found no significance found on the interaction variable coefficient for countries least affected by the crisis, thus, the changes in significance of the interaction
(2.29) (1.59) (0.14) (2.11) (1.81) (1.03) Tangibility 0.00075 -0.15182 0.00754 0.00036 -0.16197 0.01817
(0.02) (-1.29) (0.25) (0.01) (-1.35) (0.60) Constant 0.30102 0.08191 0.14379 0.29928 0.12099 0.19328
variable for countries experiencing a crisis could be caused by the crisis itself. This confirms that when there are limited sources of alternative financing, small firms use more trade credit to manage sales growth.
Table 5
Trade credit and firm growth
The dependent variable is growth of firm measured by annual growth of revenue. Size dummy is constructed for small firms by taking the 25th percentile of log of sales. Bank loans are scales by total
liabilities. NTC ratio is measured using equation x. The interaction variable is constructed by multiplying NTC with the size dummy. The standard errors are obtained using clustering on the country and time as explained in the paper. Absolute values of robust t-stats are in brackets
Emerging Developed
Prior During Post Prior During Post
Growth lagged -0.00253 -0.12188 -0.00275 -0.00206 -0.12463 -0.00231 (-2.36) (-2.23) (-2.39) (-2.33) (-2.09) (-2.37) Bank loans -0.00919 0.22173 -0.01367 0.00116 0.30104 -0.00633 (-0.58) (2.57) (-0.80) (0.21) (3.73) (-0.75) Size dummy -0.04893 -0.02319 -0.02782 -0.12334 -0.260089 -0.14061 (-1.66) (-6.67) (-0.92) (-6.07) (5.49) (6.23)
Net trade credit 0.71768 0.4517 0.07613 -0.34284 0.34797 -0.32631
(5.65) (3.56) (5.71) (-0.78) (3.84) (-0.74)
Net trade credit x
Size dummy 0.45665 0.47329 0.10327 0.35442 -0.153 0.33444
(-0.57) (-1.99) (-1.96) (0.80) (-0.49) (0.76)
Constant 0.11192 1.26655 0.03701 0.13246 0.83611 0.03224
5.3 Trade credit and economy development
To further understand the use of trade credit during a crisis, we look into country-specific factors that may influence trade credit. In this study we will look into banking sector development and trade credit in emerging countries. The argument is that the supply side of the theory suggests that when bank credit decreases, trade credit will also decrease due to a lack of credit supply. In developed countries were the financial markets are more mature, there is a wider choice of alternative financing. In emerging countries, however, where the financial markets are less developed, the supply of trade credit will be highly dependent on the banking sector.
To study this, we look at the interaction variable between economy development and banking sector development. First of all, the coefficients of banking sector development on trade credit are positive for all periods of this study, however, only statistically significant during the crisis period. This could imply that during normal economy conditions, when firms are more alternative financing are available, the use of trade credit does not depend highly on bank credit. When the financial market dysfunctions, however, firm become more dependent on bank loans to supply for trade credit. Looking at the interaction variable, we can see that the positive and significant coefficient implies that the dependence of banking sector development is larger for firms in emerging countries.
Table 6
Trade credit and economy conditions
The dependent variable is growth of firm measured by annual growth of revenue. Size dummy is constructed for small firms by taking the 25th percentile of log of sales. Bank loans are scales by total liabilities. NTC ratio is measured using equation x. The interaction variable is constructed by multiplying NTC with the size dummy. The standard errors are obtained using clustering on the country and time as explained in the paper. Absolute values of robust t-stats are in brackets
Emerging Developed
Prior During Post Prior During Post
Economy development 0.0386 0.00413 0.04270 -0.4443000 -0.0004456 -0.4445000 (4.85) (5.02) (3.19) (-5.49) (-5.51) (-5.48) Market development 0.4710 0.00033 0.38400 0.4470000 0.0000000 0.4610000 (2.59) (1.32) (1.33) (2.70) (2.57) (2.74)
Bank sector development 0.3610000 0.37809 0.3610000 -0.1190000 0.0000000 -0.1170000 (-1.65) (-4.65) (-1.64) (-3.00) (-2.94) (-2.82) Inflation -0.0000113 -0.0000114 -0.0000115 -0.0000006 -0.0000002 -0.0000010 (-1.39) (-1.42) (-1.43) (-0.47) (-0.20) (-0.72) Economy development x bank sector development 0.3630000 0.9880000 0.3820000 0.0704000 0.0000000 0.0772000 (3.68) (3.81) (3.60) (1.23) (1.36) (1.31) Constant -0.0000448 -0.0000448 -0.0000458 -0.0000108 -0.0000096 -0.0000048 (-5.53) (-5.53) (-2.93) (-3.39) (-3.41) (-1.64) 6. CONCLUSION
I study the use of alternative debt financing, namely trade credit, of firm after a financial crisis. Although there is an aggregate increase in trade credit during a crisis, the results of this study shows different pattern of trade credit for different types of firms. Love et al. (2007) explains that net trade credit (NTC) measures the willingness of firms to extend credit to its customers net of credit from suppliers. To further study these heterogeneous behavior of firms toward trade credit, we first look into the relationship between size of firm and trade credit. I argue that large firms that enjoy better access to bank loans during a credit crunch will use less trade credit. The high cost of trade credit makes it an unattractive financing option for large firms. At the same time, these firms are more willing and able to extend credit to its constrained customers. The results are actually contradicting to the theories and previous findings. There was a negative correlation between the large firm size dummy and NTC during the crisis period; however, I did not find any significance for this coefficient. To further look into this, I also look into trade receivables and trade payables of large firms during the crisis period. The results show during a crisis, large firms have statistically significant positive correlation with trade receivables, which could imply that large firms are more willing to extend credit to customers during the crisis period. At the same time, I found that large firms are increasing trade payables during the crisis. One explanation of this increase in trade payables is delayed repayments from customers due to temporary illiquidity caused by the crisis causing credit to accumulate and pile up. This affect continues even after the after crisis period which is contradicting to previous findings. Murfin and Njoroge (2013) explain in their paper that large firms that have dominating power over small firms are
able to bargain favorable terms with smaller suppliers to provide trade credit. This could explain the positive relationship between large firms and trade payables in the after crisis period.
Both large firms and small firms use a wide amount of trade credit. While large firms often use trade credit as “cash management control”, small firms actually rely on trade credit for its daily business activities, especially during times of credit constraint. During times of financial contractions, trade credit can encourage sales to customers that might be financially constrained. This is more important for small firms that have limited access to the capital markets to fund sales. Therefore, we can expect small firms to have a higher reliance on trade credit to manage its sales growth. The results of this study shows that trade credit is important to firm growth in all three periods of the study. However, the impact is greater for small firms especially during the crisis period. This confirms the importance for small firms to manage growth through trade credit especially during times of limited access to external funding.
A firm’s choice of external financing is also highly dependent to market wide factors. For example, in emerging countries where the financial markets are less mature firms commonly resort to bank credit as their primary source of fund. The dependence of firms in emerging countries on bank sector development is expected to influence its behavior towards alternative external financing. The supply side of the theory suggest that when there is a bank lending constraint, trade credit is also expected to reduce due to a lack of credit supply. Furthermore, in countries where the financial markets are less developed, firms will have high dependence on bank loans to finance trading. The results of this study highlights the importance of bank credit availability, especially in time of financial crisis.
Reference List
Astrauskaite, I., & Paškevicius, A. (2014). Competition between Banks and Bond Markets: Hardly Impacted or Softly Complemented. Procedia Economics and Finance, 9, 111-119.
Berger, A. N., Klapper, L. F., & Udell, G. F. (2001). The ability of banks to lend to informationally opaque small businesses. Journal of Banking & Finance, 25(12), 2127-2167.
Blazenko, G. W., & Vandezande, K. (2003). The product differentiation hypothesis for corporate trade credit. Managerial and Decision Economics, 24(6‐7), 457-469.
Boissay, F., & Gropp, R. (2007). Trade credit defaults and liquidity provision by firms. Bolton, P., & Freixas, X. (2000). Equity, bonds, and bank debt: Capital structure and financial market equilibrium under asymmetric information. Journal of Political Economy, 108(2), 324-351.
Carey, M., Post, M., & Sharpe, S. A. (1998). Does corporate lending by banks and finance companies differ? Evidence on specialization in private debt contracting. The Journal of Finance, 53(3), 845-878.
Cole, R. A., Goldberg, L. G., & White, L. J. (2004). Cookie cutter vs. character: The micro structure of small business lending by large and small banks. Journal of financial and quantitative analysis, 39(02), 227-251.
Cuñat, V., 2002. Suppliers as debt collectors and insurance providers. FMG discussion paper Series DP365.
De Meza, D., & Webb, D. C. (1987). Too much investment: a problem of asymmetric information. The quarterly journal of economics, 281-292.
Deesomsak, R., Paudyal, K., & Pescetto, G. (2009). Debt maturity structure and the 1997 Asian financial crisis. Journal of Multinational Financial Management, 19(1), 26-42. Denis, D. J., & Mihov, V. T. (2003). The choice among bank debt, non-bank private debt, and public debt: evidence from new corporate borrowings. Journal of financial Economics, 70(1), 3-28.
Esser, K., Hillebrand, W., Messner, D., & Meyer-Stamer, J. (2013). Systemic competitiveness: new governance patterns for industrial development. Routledge.
Fan, D. K., & So, R. W. (2004). What managers think of capital structure: the evidence from Hong Kong. Journal of Asian Economics, 15(4), 817-830.
Fan, H., Lai, E. L. C., & Li, Y. A. (2013). Credit constraints, quality, and export prices: Theory and evidence from china (No. 4370). CESifo Working Paper.
Ferrando, A., & Mulier, K. (2013). Do firms use the trade credit channel to manage growth?. Journal of Banking & Finance, 37(8), 3035-3046.
Ferris, J. S. (1981). A transactions theory of trade credit use. The Quarterly Journal of Economics, 243-270.
Garcia-Appendini, E., & Montoriol-Garriga, J. (2014). Trade Credit and Financial Distress. Available at SSRN 2457750.
Goldstein, M. (1998). The Asian financial crisis: Causes, cures, and systemic implications (Vol. 55). Peterson Institute.
Honohan, P., & King, M. (2009). Cause and effect of financial access: cross-country evidence from the Finscope surveys. World Bank mimeo.
Li, B., Nguyen, V. T., & Lin, B. (2007). A quantitative analysis on the impacts of bank loans for small enterprises. International Journal of Management and Enterprise Development, 4(2), 119-127.
Love, I., Preve, L. A., & Sarria-Allende, V. (2007). Trade credit and bank credit: Evidence from recent financial crises. Journal of Financial Economics, 83(2), 453-469. Mac an Bhaird, C. (2013). Demand for debt and equity before and after the financial crisis. Research in International Business and Finance, 28, 105-117.
Murfin, J., & Njoroge, K. (2013). The implicit costs of trade credit borrowing by large firms. Available at SSRN 2023409.
Nilsen, J. H. (1999). Trade credit and the bank lending channel.
O'Toole, C., & Casey, E. (2014). Bank-lending Constraints, Trade Credit and Alternative External Finance since the Financial Crisis: Evidence from European SMEs (No. RB2014/2/3).
Pagano, P., & Schivardi, F. (2003). Firm Size Distribution and Growth*. The Scandinavian Journal of Economics, 105(2), 255-274.
Schwartz, R. A. (1974). An economic model of trade credit. Journal of Financial and Quantitative analysis, 9(04), 643-657.
Wilner, B. S. (2000). The exploitation of relationships in financial distress: The case of trade credit. The Journal of Finance, 55(1), 153-178.
