Does firm age increase selfishness?
A trade credit perspective
Marieke van de Burgwal
1Rijksuniversiteit Groningen
Thesis for the Master of Finance
First thesis supervisor: Dr. J.H. von Eije
Second thesis supervisor: Dr. W. Westerman
Abstract.
Panel regression analysis's with firms fixed-effects are conducted to investigate the relation between selfish trade credit behaviour and firm age from manufacturing firms from the United
Kingdom. I find evidence that older firms receive less and provide less trade credit than younger firms. When combining both effects in an overall measure of selfishness, I find that older firms act more selfish than younger firms. Selfish trade credit behaviour is also affected by firm size, long-term debt and net income.
Keywords: trade credit, selfish trade credit behaviour, firm age, selfishness
2 1. Introduction
There was a time when people participated in bartering; one good for the other. No money existed yet. At some point, bartering became inconvenient and even troublesome. Then coins provided the solution as a means of payment, calculating and value. Ferris (1981) explains that trade credit takes in the same intermediate position as coins took in the bartering trade. Coins and trade credit both try to smooth trade and in the end, trade credit is nothing more than a post-payment, according to Ng, Smith and Smith (1999). It is a short-term loan that comes into existence with the exchange of goods. Besides these narrow definitions of trade credit, wide definitions of trade credit in which trade credit is introduced as a strategic tool or a strategic asset are becoming more important.
Receiving trade credit increases a firm's accounts payable and providing trade credit increases a firm's accounts receivable. Most firms are receivers and providers of trade credit simultaneously. The firm receiving the trade credit is permitted to pay the money at a later date while receiving the goods at an earlier date. Though, the mechanics behind trade credit remain the same, the motives behind trade credit and their interpretation changes. Theories behind trade credit are related to motives to provide or receive trade credit instead of paying or receiving cash immediately. One literature stream focuses on more selfish motives to participate in trade credit, while another literature stream leans to more cooperative trade credit motives.
Von Eije and Tewolde (2007) elaborate on these different streams in the literature on trade credit. They suggest that firms in developing countries, surviving in a financial weaker business environment, are supposed to be more selfish and firms in developed countries, participating in a modern, relationship-based business society, are supposed to be more cooperative. Von Eije and Tewolde (2007) explain that a firm behaves more selfish when it minimizes its accounts receivable and maximizes its accounts payable. I will follow the footsteps of Von Eije and Tewolde (2007) and use the selfish and cooperative trade credit streams as a starting-point for my research.
There are indicators in the literature that some types of firms may become more selfish than others. There is no research yet that studies whether selfishness is more prevalent in some types of firms within a country. I try to start to fill in this gap by analysing whether a firm's age is a determinant of selfish trade credit behaviour in the United Kingdom by conducting empirical research.
3
receivable as well as on the combined selfishness measure. For practice it is interesting to learn how firm age effects selfish trade credit behaviour as it may help to influence decisions on providing (or receiving) trade credit to (from) firms of different maturities.
Cunat (2002) and Altunok (2011) use the variable firm age as a determinant variable in their empirical studies. Cunat (2002) investigates the relation between firm age and accounts payable, described as the level of trade credit. Altunok (2011) investigates the relation between firm age and trade credit benefits, described as the price reductions for firms that use trade credit.
Since firm behaviour is observable but intentions are not always observable, I deliberately write about trade credit behaviour instead of trade credit policy or strategy. Circumstances rather than intentions may create selfish or cooperative trade credit behaviour. It may be the case that firms have no options at hand and are forced to be cooperative or selfish. For this reason, trade credit behaviour refers to all intended and unintended firm behaviours related to trade credit. Furthermore, selfish trade credit motives may result in cooperative trade credit behaviour and cooperative trade credit motives may result in selfish trade credit behaviour.
In section two a literature background on trade credit is provided. This literature background is chronologically ordered. The third section is a description of the data and methodology. The results are presented in the fourth section. The conclusion is provided in the fifth section.
2. Literature background
Schwartz (1974) introduces selfish trade credit motives by stating that credit providers intend to maximize their profits by providing trade credit to their customers. This results in cooperative trade credit behaviour, since this increases a firm's accounts receivable. Schwartz (1974) identifies two selfish motives to act cooperative, namely the transaction and the financing motive.
According to Schwartz (1974), the transaction motive comes from the desire to simplify credit
receivers' cash management, make more accurate forecasts of future outlays and to make plans for unexpected purchases. Transaction costs are reduced since matching the time pattern of payments with the time pattern of delivery of goods is costly. Bills are not paid and cumulated to create payments on a regular basis. This is especially worthwhile when goods are delivered irregular or at a high frequency. At the same time, the trade credit receivers benefit from extra financial resources and the related time value of money. For example, a customer may buy goods for 100 euro at day 0. The customer and the supplier agree that the customer pays the bill of 100 euro at day 30. In those 30 days, the customer has still full access to its 100 euro which it has not yet spend on the goods sold.
The financing motive explains that some firms have easier and cheaper access to capital
4
money to provide trade credit to customers. Financial institutions would not provide credit as easy and cheap to these trade credit receiving firms. Poor collateral and higher levels of risk increase the probability of default. Due to this probability of default, financial institutions are less willing to provide trade. Trade credit providing firms experience a better collateral and lower levels of risk with the same customer than financial institutions, since they are in a better position to repose the goods sold, resell them, and also better able to monitor their customers.
Especially firms with productive investment opportunities are considered to be appropriate
candidates to receive trade credit for trade credit providers. Schwartz (1974) concludes that especially established firms help young firms with productive investment opportunities through providing trade credit. Older firms perceive the promising young firms as an investment opportunity and try to commit the young firms to their business. Schwartz (1974) also suggests that trade credit might substitute for bank credit in times of monetary restraint and credit rationing. In the current financial crisis, this phenomenon is witnessed because banks are less willing to provide credit. Due to poor economic conditions, banks have become more risk-averse and try to avoid lending money; in particular to high default risk customers. This could thus make trade credit an interesting source of finance for the latter group.
Ferris (1981) focuses on the transaction motive behind trade credit, but from a different perspective than Schwartz (1974). Ferris (1981) explains trade credit as a separation of the exchange of goods in time and place from the exchange of money. The flow of goods remains the same, while the flow of money can be adjusted to reduce uncertainty. The payments are made at a later date while the goods are received at an earlier date.
Firms do not order goods at a regular pace or quantity, which causes uncertainty in the flow of
goods. As a domino effect, this creates uncertainty in the flow of payments. According to Ferris (1981) this will lead to precautionary money holdings. The higher the variability in the flow of money, the more money a firm will hold. Trade credit does not prevent firms from holding money. Bills need to be paid before the end of the credit period and money needs to be hold to ensure these payments. A firm receiving trade credit also experience variability in the flow of money. The difference between a trade credit receiving firm and a firm paying with cash immediately is the level of uncertainty. A firm paying with cash needs to hold excess amounts of money to compensate for a higher level of uncertainty of the flow of money. A firm receiving trade credit only holds the amount of money equal to payments to be made. This lowers the costs for holding money for a trade credit receiving firm in comparison to a firm paying with cash immediately.
Ferris (1981) states that holding money comes at a cost, namely forgone interest. The interest
5
money has not been invested. The higher the amount of money held, the higher the forgone interest is. Ferris (1981) explains that trading partners jointly participate in trade credit to attempt to reduce money inventory holding costs. Hereby, Ferris (1981) identifies the awareness among trading partners to mutually benefit from trade credit. This awareness initiates cooperative behaviour, though this behaviour is solely related to money inventory holding costs. Where Schwartz (1974) explains the transaction motive from a selfish perspective, Ferris (1981) thus identifies a cooperative element though the same motive can also be interpreted from both a selfish perspective as a cooperative perspective.
In line with Schwartz (1974), Ferris (1981) states that trade credit is an investment decision.
Such a decision is conducted when the present value of its benefits exceed the present value of its costs. As a trade credit benefit, Ferris (1981) identifies lower money inventory holding costs for both trading partners, but Ferris (1981) also considers higher information and enforcement costs as trade credit costs. These benefits and costs are compared with the usage of immediate cash or with using a bank loan instead of trade credit.
A bank is specialized in providing loans and intermediates financially between firms. For this
reason, a bank can provide credit at a low cost, according to Ferris (1981). Though, a bank will only evaluate a firm's individual variability in money held. Due to precautionary money holdings, the sum of individual variability is higher when no trade credit is used. Trade credit diminishes the cash holdings of two trading partners. For this reason, the costs of the bank loan will be higher than the costs of trade credit. Therefore, Ferris (1981) states that trading partners will prefer trade credit as long as the cost is less than the sum of the individual costs of bank credit.
Ferris (1981) warns that trade credit needs to stay a source of trade credit and that it should not
become a source of financing, for both trading partners to perceive mutual benefits. This can be achieved by keeping the credit period short. When the credit period is too long, the credit providing firm will not experience trade credit as a money inventory holding costs reducing mechanism. Ferris (1981) states that in such a case too much forgone interest is missed and trade credit will be perceived as an illiquid asset to the trade credit providing firm. However, the trade credit receiving firm will perceive trade credit as a convenient source of financing when the credit period is longer. Then, joint benefits will decline and the costs will rise with the length of the credit period.
6
Another solution then lies in providing trade credit. According to Emery (1984), this solution
is less costly than adapting a firm's operations to demand fluctuations. He suggests that temporary relaxing or tightening credit terms or credit standards attracts (or pushes off) customers buying goods on credit. Instead of changing a firm's operations, demand is changed. The capacity of a firm's operations is not used to match demand but the capacity from accounts receivable is used to change demand. In this way, a firm's operations are not tied to demand and the firm maintains its operating flexibility. For this reason, Emery (1984) calls this the pure operating flexibility motive for providing trade credit.
In line with Ferris (1981), Emery (1984) indicates that firms need to hold a certain liquid
reserve to be able to meet payments and avoid excess borrowing or even insolvency costs. This liquid reserve can be held in many different forms, as long as it remains liquid or can be liquidated in a very short period of time. Emery (1984) explains that the liquidity reserve can be used in demand deposits and by the purchase of securities, like government bonds. An alternative usage for this liquidity reserve is providing trade credit, according to Emery (1984). Hereby, Emery (1984) stretches even more than Ferris (1981) the usage of trade credit as a strategic asset, an investment decision that creates a return by reducing operating costs.
Where Ferris (1981) warns trade credit should not become a source of finance, Emery (1984)
identifies a financial motive in providing trade credit. This financial motive becomes evident when the firm providing trade credit can charge an implicit interest rate that is greater than the market rate of return. Firms then act as a financial intermediary. For this reason, Emery (1984) calls this the pure financial intermediary motive for providing trade credit. The pure financial intermediary motive comes from the desire to earn an excess rate of return or, equivalently, a positive net present value on the liquidity reserve, according to Emery (1984). The market rate is the interest rate paid for borrowing money from financial institutions. The implicit rate is the rate implicitly charged when providing trade credit. Emery (1984) suggests that firms borrow money at the market rate from financial institutions and, in turn, provide trade credit to customers of the firm. As stated before, trade credit is a short-term loan. Though, the interest paid over this loan is not explicitly agreed upon or evident in the credit period. Customers are mostly unaware of the implicit rate charged on the trade credit they receive. An example is provided in appendix 1.
Emery (1984) elaborates on the financing motive of Schwartz (1974), who states that firm can
7
evident when firm's borrowing capacity is low. This is generally more the case with young firms, who Schwartz (1974) identifies as having a lower borrowing capacity.
Smith (1987) introduces trade credit as a source of information for the trade credit providing firm. Credit terms are established as a screening device to provide information about the default risk of customers. Smith (1987) indicates that high implicit interest rates facilitate the sorting of low from high default risk customers. In contrast with Emery (1984), Smith (1987) assumes awareness among trade credit receiving firms about the implicit rate charged by the trade credit providing firm.
Firms use payment discounts as an incentive for trade credit receiving firms to pay off the
trade credit more early. Standard trade credit terms state for example "net 30" or "net 60", indicating the length of the credit period. For customers this means that the full amount has to be paid within 30 days (net 30) or 60 days (net 60). Without any incentive, the trade credit receiving firm will use this short-term loan for the full length of the credit period. However, trade credit terms with trade credit options (or discounts) exist and state for example "1/10 net 30", indicating a 1% discount on the payment when the trade credit receiving firm pays within 10 days instead of the 30 days. The firm is allowed to pay within 30 days but will only receive a discount when it pays for the goods bought within 10 days. This discount is an incentive for the trade credit receiving firm to pay for the goods bought earlier. An example is given in appendix 1.
The reason for offering discounts is that the trade credit providing firm then can screen
between customers paying within 10 days or 30 days. Paying for the goods bought within 10 days indicates a low default risk customer and paying within 30 days but after 10 days indicates a high default risk customer. Paying late (between 10 and 30 days) means that the customer is willing to borrow at the high implicit rate. For the trade credit providing firm this proves that the customer has no other lower cost financing sources. According to Smith (1987), this customer is then already screened by financial institutions as a high default risk customer. These may, in general, be relatively young firms and older firms in distress. Hereby Smith (1987) identifies older and younger firms as trade credit dependent and therefore as selfish.
The trade credit providing firm is willing to pay a price for receiving information on customer
default risk. This price is the difference between the money received by the firm providing trade credit without a discount minus the money received for the goods sold with a discount for early payment. This price is also paid for the fact that money is received earlier and can be dedicated to other investment opportunities, such as providing trade credit to another customer.
8
bought. If the quality of the goods differ from the trade credit receiving firm's expectations, the goods are returned and no payment is made.
In contrast with regular product warranties, the trade credit receiving firm is not required to
objectively verify the difference between the quality received and the quality promised, according to Lee and Stowe (1993). For some products this is very difficult, if not impossible. Especially when quality issues are confused with preferences. Firms may want to return goods bought because these goods are not in line with the preferences or taste of their customers. This has not necessarily anything to do with the quality of the products and it may even be related to estimation errors of the trade credit receiving firm.
Lee and Stowe (1993) state that firms with no reputation, which are generally young and small
firms, are more likely to use trade credit rather than product warranties. Young and small firms act selfishly by preferring trade credit to test the goods bought. Due to a lack of commitment, younger firms are more imprudent to return the goods bought than established firms with developed business relations.
The trade credit providing firm can be categorized into lower quality producers and higher
quality producers. Lee and Stowe (1993) indicate that lower quality producers intend to offer larger cash discounts to induce buyers to pay with cash rather than on credit. Paying with cash increases the risk of poor quality goods for the buyer. If the quality of the goods turn out to be poor, the buyer needs to put time and effort in objectively verifying this poor quality. Furthermore, the buyer may not succeed in verifying poor quality. Moreover, the buyer solely bears the risk of a wrong estimation of customers' preferences. Lee and Stowe (1993) state that higher quality producers offer lower cash discounts because they are more certain that their products will not be perceived as poor quality products. Therefore, lower quality producers act more selfish since they are less willing to provide trade credit than higher quality producers.
In line with Smith (1987), Petersen and Rajan (1994) indentify trade credit as a price-discrimination mechanism. Both, Schwartz (1974) and Smith (1987) argue that high default risk firms will use trade credit instead of other sources of finance because they are unavailable or too expensive. Petersen and Rajan (1994) state that low default risk firms can secure other sources of finance at better terms from financial institutions. These firms perceive trade credit as more expensive since the implicit rates charged are higher than the interest rates charged on bank loans. This results in price discrimination since low default risk customers will pay immediately (and therefore use a bank loan) and high default risk customer want to use trade credit. This is in line with Brennan, Maksimovic and Zechner (1988), who identify price discrimination between cash and credit customers. The cash customers have a low risk of default and the credit customers have a high risk of default.
Petersen and Rajan (1997) suggest that small firms use more trade credit but for a different
9
access to credit from financial institutions. In line with Schwartz (1974), they suggest that suppliers with better access to credit are willing to provide trade credit to dependent firms. This suggest that small (and generally younger) firms are trade credit dependent and large (and generally older) firms are trade credit independent. Small (and young) trade credit dependent buyers then match with large (and older) financially healthy suppliers, whereby the small firms acts selfishly and the larger firms act cooperative.
Trade credit providing firms experience a competitive advantage when providing trade credit.
Goods sold with trade credit function as an implicit equity stake. These goods can be reposed and sold
to another customer. Like Smith (1987), Brennan, Maksimovic and Zechner (1988) and Lee and
Stowe (1993), Petersen and Rajan (1997) consider that trade credit also provides firms with information signals. These signals are related to the size and timing of the customer's orders and payments. A reduction in the size of customer orders and later than normal payments are both signals for the supplier that the customer is doing not so well. As a response to negative signals the supplier may start to monitor the trade credit receiving firms more closely. This is also underlined by Schwartz (1974).
However, Petersen and Rajan (1997) find also that negative income and losses in sales of the
underperforming firm may amazingly even foster the willingness to provide trade credit. And they find that trade credit is increased in times of financial distress. They interpret this cooperative behaviour as resulting from the will to stay in business (even with the help of relative risky and underperforming clients). In this case trade credit enables firms to buy goods while they would otherwise not be able to. This means that the number of sales and the customer base increases when firms provide trade credit. In line with Schwartz (1974), Petersen and Rajan (1997) find that suppliers are more willing to provide trade credit when they believe buyers have potential growth opportunities. Petersen and Rajan (1997) state that firms consider the present value of current and future margins. Trade credit receiving firms can show poor performances but this may be temporarily. This shows that Petersen and Rajan (1997) consider trade credit as resulting from a long-term commitment to an endurable buyer-supplier relationship. This commitment is based on expectations and anticipations of future business.
10
In line with Schwartz (1974), Nilsen (2002) argues that also larger (and generally older) firms use more trade credit as a source of finance in times of tight monetary policy. Schwartz (1974) and Petersen and Rajan (1993) argue that larger (and generally older) firms can find other sources of finance than trade credit. However, Nilsen (2002) identifies that larger firms without bond ratings, who lack collateralized assets use trade credit similar to smaller firms, even if they are cash rich. Larger firms without bond ratings have a lower creditworthiness and are unable to borrow from financial institutions in times of financial distress. Therefore, Nilsen (2002) suggests that also larger firms can be credit constraint and therefore can dependent on trade credit. This indicates that both larger and smaller firms are selfish by increasing their accounts payable in times of tight monetary policy. In times of tight monetary policy, financial institutions are less willing to provide bank credit. This indicates that selfish trade credit behaviour may be influenced by access to other sources of finance. Nilsen (2002) states that larger firms with a bond rating, do not use trade credit because they have alternative sources of finance, even when monetary policy is restrained.
Nilsen (2002) anticipates that the risk-aversion of financial institutions is effected by monetary
policy. This does not have to be the case. In the financial crisis, started in 2008, a stimulating monetary policy became ineffective. Though the European Central Bank keeps interest rates very low, banks are still reluctant to lend money to firms. This implicates that the risk-aversion of financial institutions influences a firm's access to credit, rather than monetary policy. Therefore, the effectiveness of monetary policy may have implications for a firm's borrowing capacity.
Banerjee, Dasgupta and Kim (2004) also suggest that the importance of the customer plays a role in providing of trade credit. A principal customer is an important customer to the supplier since this customer is responsible for a substantial part of the supplier's sales. Banerjee, Dasgupta and Kim (2004) state that these customers can use their bargaining power to delay payments. This selfish behaviour may pay off because Wilner (2000) notes that the majority of the suppliers do not charge a late payment penalty, especially when the supplier has no other substitutes for the customer.
It can be expected that younger firms have more principal customers because they their
customer base is still small. Therefore each customer has more power over the supplier and is responsible for a substantial part of the younger firm's sales. The younger firm may have no other option than to be cooperative and to supply more trade credit by extending the credit period. At the same time, it is reasonable that the younger firm will not charge late payment penalties to prevent its customer to go to another supplier. Both aspects suggest that younger firms are less selfish.
In line with Petersen and Rajan (1997), Banerjee, Dasgupta and Kim (2004) identify that
11
their assets and liabilities, and therefore use trade credit as a strategic asset to minimize their risk.
2.1 Empirical research with firm age
Cunat (2002) investigates the relationship between accounts payable and firm age. He conducts a panel regression and identifies a non-linear relationship between the level of trade credit over assets and the age of a firm. The first three years of a firm's life, a positive relation is found and this relations becomes negative from the fifth year of a firm's life. He indicates that other empirical studies found negative linear relations between trade credit and firm age. Cunat (2002) explains that the level of trade credit grows with the relation build between a supplier and its customers. He also argues that trade credit entails that suppliers have a stake in the survival of their customers. Petersen and Rajan (1997) and McMillan and Woodruff (1999) underline this by identifying trade credit as a long-term commitment and a sign of trust. Cunat (2002) suggests that a growth in retained earnings explains the negative relations between firm age and accounts payable. He argues that firms then no longer depend on expense trade credit as a source of finance.
Altunok (2011) suggests that larger, older and more creditworthy firms are offered more trade credit. She finds that larger, older, more creditworthy firms benefit the most from trade credit. This benefit can be regarded as the price reduction for firms that use trade credit. These price reductions are related to the various trade credit terms. This is in line with Petersen and Rajan (1994) who state that low default risk customer, who are more creditworthy, can secure better credit terms. This is also in line with Banerjee, Dasgupta and Kim (2004) who suggest that young firms provide more trade credit by extending the credit period. This suggests that older firms have more bargaining power and therefore are more selfish.
2.2 Empirical research with selfishness
12 2.3 Selfishness and firm age
Some of the literature suggests that younger firms are more selfish than older firms (Schwartz (1974), Emery (1984), Lee and Stowe (1993) and Petersen and Rajan (1997)). Younger firms do not as easily and cheaply attain credit as older firms. Therefore, they often depend on trade credit provided by older firms, whereby trade credit functions as the only source of finance. Older firms are willing to perform the role of a financial intermediary when these younger firms have potential to grow. This also has to do with the advantages older firms have over financial institutions when providing credit. Older firms are in better position to receive information signals more timely through trade credit. Therefore, they are better able to screen customers through the customer's use of implicit interest rates, discounts and late payments. Deviations from normal buying activities can also provide information signals. At the same time, older firms can more easily reposes and resell the goods sold than financial institutions. Furthermore, trade credit creates certainty in the flow of money. This also results in a reduction in the sum of individual variability of money held which reduces forgone interest. For all these reasons, trade credit providing firms experience better collateral and lower risks of default than financial institutions do when extending credit to the same customers.
Younger firms have less reputation than older firms. This lack of reputation induces firms to
use trade credit rather than product warranties, as a period to test the quality of the goods. Another convenience is that money is not yet transferred. This increases the power of the younger firm to send the goods back and reverse the transaction. At the same time, the younger firm can use the money not yet spend on the goods bought, for other purposes. This also increases the desire to extend the credit period. For young finance-constrained firms that rely on new investments, this may be very beneficial.
Most literature implies that young firms are selfish because they receive more trade credit,
which increases their accounts payable. Emery (1984), thus, suggests that younger firms are more selfish because they are less likely to provide trade credit. He states that younger firms have a low borrowing capacity, which implies that they are less likely to provide trade credit since they cannot borrow money from financial institutions.
13
competition. Therefore, the younger firm provides more trade credit when their customers demand a longer credit period. This cooperative trade credit behaviour increases the younger firm's accounts receivable since they provide more trade credit by extending the credit period. Moreover, older firms may not only be offered more trade credit but also at more beneficial trade credit terms. Furthermore, financial distress can also make older firms, who lack collateralized assets, credit-constraint and use therefore more trade credit. This suggests that older firms are highly selfish, not only by receiving more trade credit but also through beneficial trade credit terms.
These literature stances are thus not conclusive on the question whether younger firms or older firms are more selfish. This investigation aims to shed light on this issue. The main research question underlying this paper is then as follows:
Are firms more selfish when they are older?
Previous literature shows that older firms may be more as well as less selfish than younger firms. This suggest the following set of hypothesises:
Ho: Older firms are equally selfish than younger firms.
H1: Older firms do act more (or less) selfish than younger firms.
Besides selfishness measures, the firm's age is used to test the hypothesises.
3. Data and methodology
Manufacturing firms operating in the United Kingdom are selected because the United Kingdom is a large country with a rich financial history and trade credit is not customary with many service firms. As banks and insurance companies do not regularly provide trade credit either, financial institutions are also omitted. Finally, also utilities are discarded because of their special public good characteristics.
Data from accounts payable, accounts receivable, total assets, long-term debt and net income are collected for the years 2003 until 2012. The year of incorporation is also collected for every firm. Panel regression analysis's with firm fixed-effects are conducted to investigate the relation between firm age and selfish trade credit behaviour. The significance level for each regression analysis is 5%. This gives the equation:
SELFISHNESSit = α + β1AGE it + εit (1)
The independent variable AGE is calculated for each firm (i) as the panel year (t) minus the year of
incorporation (YINCi). The independent variable AGE is calculated with the formula
2
:
2
14
AGEit = t - YINCi, with t = 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012.
For the dependent variable SELFISHNESS I take the log of the ratio of accounts payable (AP) to total assets (TA) plus one minus the log of the ratio of accounts receivable (AR) to total assets (TA) plus one. The dependent variable SELFISHNESS is thus expressed in the formula:
SELFISHNESSit = ln((AP/TA)it + 1) − ln((AR/TA)it + 1)
To investigate the relation between firm age and selfish trade credit behaviour, the results need to be controlled for other confounding variables and therefore firms from one sector, in this case the manufacturing sector is selected. The following variables are also created to enable a more accurate prediction of the relationship between firm age and selfish trade credit behaviour.
Schwartz (1974), Lee and Stowe (1993) and Petersen and Rajan (1997) indicate that smaller
firms receive more trade credit than larger firms. This indicates that smaller firms may be more selfish than larger firms. However, Altunok (2011) and Banerjee, Dasgupta and Kim (2004) suggest that larger firms have bargaining power and thus receive more trade credit. Also Nilse (2002) indicates that larger firms may be credit-constraint and depend on trade credit. This indicates that larger firms are more selfish. The variable SIZEt-1 is created to control for firm size. The variable SIZEt-1is the one
year lagged-value of the log of total assets (TA).3 The formula for this variable is: SIZEi,t-1 = ln(TA)i,t-1.
Banerjee, Dasgupta and Kim (2004) suggest that less leveraged firms can afford to provide more trade credit. This indicates that highly leveraged firms are more selfish than firms with less leverage. Schwartz (1974) gives a contrasting perspective. He explains that firms provide more trade credit when they have easier and cheaper access to financial resources from financial institutions. Leverage creates willingness to provide trade credit, according to Schwartz (1974). Emery (1984) agrees with Schwartz (1974) and suggests that firms profit from using their borrowing capacity to exploit this capacity by extending trade credit. The variable DEBTt-1 is created to control for firm
leverage. This variable is the one year lagged-value of the ratio of long-term debt (LTD) to total assets
(TA). The formula corresponding to the debt variable is: DEBTi,t-1 = (LTD/TA)i,t-1.
Firms do not dependent on external sources of finance when they have internal sources of
finance. An important internal source of finance is net income, which can be reinvested, used for dividends or used to pay creditors and to allow accounts receivable to be increased. A firm is expected
to be more selfish when it has less net income and so the variable NET INCOMEt-1 is created. This
variable is the one year lagged-value of the ratio of net income (NI) to total assets (TA) and is
provided with the formula: NET INCOMEi,t-1 = (NI/TA)i,t-1.
The variable RISKt-1 describes the uncertainty a firm needs to bear given its net income.
Uncertainty in net income increases the urge to find other sources of finance. Risk also increases the
3
All the independent financially base variables (SIZEt-1, DEBTt-1, NET INCOMEt-1 and RISKt-1) are lagged with
15
probability of default. Both increase the need for trade credit and to become more selfish. Therefore, the variable RISKt-1 is included. The variable RISKt-1 is calculated as the standard deviation of the
ratio of net income to total assets over three years. The variable risk is calculated with the formula: RISKi,t-1 = σ3(NI/TA)i,t-1 with σ3 = standard deviation of the net income to total assets over three years.
Including the variable RISKt-1 has a consequence for the data retrieved for the other variables.
The lagged standard deviation of net income to total assets cannot be calculated for the years 2003, 2004 and 2005. Therefore, only the period from 2006 until 2012 can be used for all variables in panel analysis.
Dummy variables are created for the years 2007, 2008, 2009, 2010 and 2011. These
year-dummies variables are expressed as D07, D08, D09, D10 and D11, respectively. The year-year-dummies are used in all regressions but are not shown in the equations. In times of financial turmoil, firms find it more difficult to attain credit from financial institutions. Nilsen (2002) indicates that tight monetary policy may influence the access of other sources of finance. This induces selfish trade credit behaviour, since firms will use trade credit more as a last source of finance.
The main relationship extended with confounding variables gives the equation:
SELFISHNESSit = α + β1AGEit + β2SIZEi,t-1 + β3DEBTi,t-1 + β4NET INCOMEi,t-1 + β5RISKi,t-1 + εit (2)
The descriptive statistics of the sample to investigate the relation between firm age and selfish trade credit behaviour is provided in table 1.
[Please insert Table 1 here]
The relationship between firm age and selfish trade credit behaviour has three possible results; no relation, a positive relation or a negative relation. A positive relation indicates that older firms are more selfish. A negative relation indicates younger firms are more selfish. If there is no relation I am unable to explain whether older or younger firms are more selfish. Such an explanation is, however, based on the interaction between SELFISHNESS and AGE in equation (1). In table 2 possible interactions between these variables are provided whereby SELFISHNESS is presented as the difference between ln(AP/TA + 1) and ln(AR/TA+1). Therefore, the interaction between ln(AP/TA + 1), ln(AR/TA+1) and firm age result in four possible positive relations and four possible negative relations.
[Please insert Table 2 here]
16
payable is also investigated separately from the relation between firm age and accounts receivable.
The dependent variable ln(AP/TA + 1) is the log of the ratio of accounts payable (AP) to total
assets (TA) plus one. The dependent variable ln(AP/TA + 1) investigates the relation between firms age and accounts payable. This gives the equation:
ln(AP/TA + 1)it = α + β1Age it + β2SIZEi,t-1 + β3DEBTi,t-1 + β4NET INCOMEi,t-1 + β5RISKi,t-1 + εit (3)
The dependent variable ln(AR/TA+1) is the log of the ratio of accounts receivable (AR) to
total assets (TA) plus one. The dependent variable ln(AR/TA+1) investigates the relation between firms age and accounts receivable. This gives the equation:
ln(AR/TA+1)it= α + β1Age it + β2SIZEi,t-1 + β3DEBTi,t-1 + β4NET INCOMEi,t-1 + β5RISKi,t-1 + εit (4)
A robustness-test is conducted to test the robustness of the SELFISHNESS model. The dependent variable SELFISHNESS is replaced with the dependent variable SELFISHNESS-CHECK. The variable SELFISHNESS-CHECK is the log of the ratio accounts payable to accounts receivable plus
one. The formula corresponding to this variable is: SELFISHNESS-CHECKit = ln((AP/AR)it + 1).
The descriptive statistics of the sample to investigate the relation between firm age and accounts receivable, firm age and accounts payable and a robustness-test are provided in table 3.
[Please insert Table 3 here]
From table 3 it may be noticed that the number of observations is much less with the sample for the variable SELFISHNESS-CHECK, which is caused by many zero accounts payables and many zero accounts receivables.
4. Results
4.1 Correlation matrix
A correlation matrix is conducted to identify to what extend the independent, control and dummy variables correlate with each other. A correlation below -0.7 or above 0.7 indicates a high level of correlation. Then two variables are co-linear. This may negatively influence the results of the regression analysis's. Table 4 provides the correlation matrix.
[Please insert Table 4 here]
17 4.2 Firm age and selfishness
The relation between accounts payable and firm age is investigated to find if firms receive more or less trade credit when firm age increases. This relation is investigated with the dependent variable ln(AP/TA + 1). The coefficient of the variable AGE in this model will indicate which relations in table 2 are possible results.
The relation between accounts receivable and firm age is investigated to find if firms provide
more or less trade credit when firm age increases. This relation is investigated with the dependent variable ln(AR/TA+1). The coefficient of the variable AGE in this regression will indicate which relations in table 2 are possible results.
To determine which relation in table 2 (relation 1 until 8) is evident, the relation between selfishness and firm age is investigated with the variable SELFISHNESS. The results for the dependent variables ln(AP/TA + 1), ln(AR/TA+1) and SELFISHNESS are provided in table 5.
[Please insert Table 5 here]
The panel regression model ln(AP/TA + 1), presented in the first column of table 5, indicates that the
independent variable AGE is significant. The control variables SIZEt-1 and NET INCOMEt-1 are also
significant. However, the control variables DEBTt-1 and RISKt-1 are not significant.
The negative coefficient of the variable AGE4 indicates that the ratio of accounts payable to
total assets decreases when firms are ageing. This corresponds with the decreasing sign (- or --) in the relations 4, 5 and 6 in table 2 for ln(AP/TA + 1). This indicates that older firms have lower levels of accounts payable to total assets than younger firms. This implies that younger firms are more selfish since they have higher levels of accounts payable to total assets by receiving more trade credit than older firms.
Furthermore, the negative coefficient of the variable SIZEt-1 indicates that smaller firms
receive more trade credit. This is in line with Schwartz (1974), Lee and Stowe (1993) and Petersen and Rajan (1997) who state that smaller firms receive more trade credit. The positive coefficient of the
variable NET INCOMEt-1 indicates that firms with a higher ratio of net income to total assets, receive
more trade credit. This is not in line with Schwartz (1974), Lee and Stowe (1993) and Petersen and Rajan (1997), who suggest that smaller firms are more trade credit dependent since these firms are unable to find other sources of finance. This suggests that firms with low incomes are more trade credit dependent and will therefore act more selfish by attaining more trade credit. Though, Schwartz (1974) mentions that trade credit includes the risk of financing customers with a higher risk of default.
4 A panel regression analysis with firm fixed-effects is conducted to investigate whether the relation between
firm age and selfish trade credit behaviour is linear. The variable AGE2it = AGE*AGE is created to investigate
18
It can be reasoned that firms providing trade credit may demand a certain income stability from firms that receive trade, to find assurance that bills can be paid for at the end of the credit period.
The panel regression model ln(AR/TA+1), in the second column of table 5, indicates that the
independent variable AGE is significant. Also, the control variables SIZEt-1 and DEBTt-1 are
significant. However, the control variables NET INCOMEt-1 and RISKt-1 are not significant.
The negative coefficient of the variable AGE indicates that the ratio of accounts receivable to
total assets decreases when firms are ageing. This corresponds with the decreasing sign (- or --) in the relations 3, 4 and 5 in table 2 for ln(AR/TA+1). This indicates that older firms have lower levels of accounts payable to total assets than younger firms. This implies that older firms are more selfish since they have lower levels of accounts payable to total assets by providing less trade than younger firms.
Furthermore, the negative coefficient of the variable SIZEt-1 indicates that larger firms provide
less trade credit. The positive coefficient of the variable DEBTt-1 indicates that firms who are less
leveraged, provide less trade credit. This is in line with Schwartz (1974) and Emery (1984). Schwartz (1974) explains that leverage creates willingness to provide trade credit. Emery (1984) underlines this by stating that firms profit from using their borrowing capacity to exploit this capacity by extending trade credit.
The third column in table 5 presents the panel regression model SELFISHNESS. The dependent
variable AGE is significant. The control variables SIZEt-1, DEBTt-1 and NET INCOMEt-1 are
significant. This model shows that the variable RISKt-1 is not significant.
The positive coefficient of the variable AGE indicates that older firms are more selfish than
younger firms. This implies that the model SELFISHNESS is effected more by the model ln(AR/TA+1) than the model ln(AP/TA + 1). With a coefficient of -0.0056 the model ln(AR/TA+1) is effected more by firm age than the model ln(AP/TA + 1) with a age coefficient of -0.0032. With every increase in firm age the model ln(AR/TA+1) predicts that accounts receivable will decline with 0.0056. The model ln(AP/TA + 1) predicts that the accounts payable will decline with 0.0032 with every increase in firm age. Therefore, accounts receivable will decline faster than accounts payable when firms are ageing. This corresponds with relation 4 in table 2.
The negative coefficient of the variable SIZEt-1 indicates that smaller firms are more selfish
because they receive more trade than they provide trade credit. As indicated before, this is in line with Schwartz (1974), Lee and Stowe (1993) and Petersen and Rajan (1997) who state that smaller firms
receive more trade credit. Furthermore, the negative coefficient of the variable DEBTt-1 suggests that
less leveraged firms are more selfish by providing less trade. As stated before, this is in line with Schwartz (1974) and Emery (1984) who indicate that less leverage causes firms to provide less trade
credit. The positive coefficient of the variable NET INCOMEt-1 indicates that firms with a higher ratio
19 4.3 Robustness-check
To check whether the panel regression model SELFISHNESS is robust, a robustness-test is conducted. The variable SELFISHNESS-CHECK tests the robustness of the model SELFISHNESS. The main concern of the robustness-test is a similar prediction of the relation between selfish trade credit behaviour and firm age as in the SELFISHNESS model. The results for the robustness model SELFISHNESS-CHECK are provided in table 6.
[Please insert Table 6 here]
The SELFISHNESS-CHECK model indicates that the independent variable AGE is significant. The
control variable RISKt-1 is also significant. However, the control variables DEBTt-1, NET INCOMEt-1
and SIZEt-1 are not significant. The AGE coefficient indicates a positive relationship between firm age
and the ratio of accounts payable to accounts receivable. The AGE coefficients in SELFISHNESS model and the SELFISHNESS-CHECK model are similar to each other, with an AGE coefficient of 0.0022 in the SELFISHNESS model and an AGE coefficient of 0.0035 in the SELFISHNESS-CHECK model. This indicates that the SELFISHNESS model is a robust model to predict the relation selfish trade credit behaviour with the firm age.
4.4 Trend implied by year-dummies
The year-dummies D07, D08, D09, D10 and D11 are not shown in the results, though they are incorporated in all four panel regression analysis's conducted. Table 7 provides the year-dummies variables for the dependent variable SELFISHNESS.
[Please insert Table 7 here]
As table 7 shows, the dummies for the years 2008, 2009 and 2010 are significant, while year-dummies for the years 2007 and 2011 are not. The coefficients of D08, D09 and D10 are, respectively, 0.0057, 0.0053 and 0.0036. This may indicate that the financial crisis, which started in 2008, has an influence on the results of the panel regression analysis's. Schwartz (1974), Smith (1987) and Nilsen (2002) underline this by indicating that tight monetary policy or poor access to financial credit increases the need for both smaller (in general younger) and larger (in general older) firms to use trade credit as an alternative source of finance.
5. Conclusions
20
selfish trade credit behaviour. A firm acts selfishly when it provides less trade credit or receives more trade credit.
21
(2004) suggest that younger firms can be forced to be cooperative with their customers. It is likely that younger firms have a small customer base, with each customer responsible for a substantial part of sales. This induces a great dependence on each customer for the young firm to stay in business, which increases the bargaining power of customers. Customers can then demand favourable credit terms. The younger firm may be anxious to lose a big customer to the competition and will therefore fulfil their demands. It is likely that these big customers are older firms with more relationships than the younger firm. This is in line with Altunok (2011), who states that older firms benefit more from trade credit and Smith (1987), who suggests that firms can benefit from trade credit when paying earlier and therefore receiving a payment discount.
It can be sub concluded that both, younger and older firms, act selfishly. Younger firms receive more trade credit while older firms provide less trade credit. A third regression analysis is therefore conducted to investigate what the overall combined effect is. A significant positive relation is found with a firm age coefficient of 0.0022. This indicates that older firms are overall more selfish than younger firms, which approves the corresponding alternative hypothesis. The hypothesis that older firms are equally selfish than younger firms is rejected, as well as the alternative hypothesis that older firms are less selfish.
On overall it can be concluded that older firms do participate less in trade credit than younger
firms. Older firms provide and receive less trade credit, while younger firms provide and receive more trade credit. This is in line with McMillan and Woodruff (1999) and Petersen and Rajan (1997) who identify trade credit as a sign of trust and a long-term commitment. This requires both trading partners to equally participate in trade credit to achieve mutual trade credit benefits as Ferris (1981) describes. Though, trade credit entails taking risks. Younger firms take these risks since they have less opportunities to attain other sources of finance. At the same time, they are eager to build endurable business relations and commit customers to their business. Older firms do not dependent on trade credit like younger firms do. They have other sources of finance, as well internal as external, to use instead of trade credit. This reduces the willingness to bare the risks related to providing trade credit.
Altunok (2011) indentified that older firms benefit more from trade credit than younger firms.
In accordance with Smith (1987), this suggests that older firms make more use of payment discounts. Then the credit period is shorter and accounts payable decreases. At the same time, benefits increase due to received payment discounts.
Limitations and advice
22
further research, it is therefore interesting to investigate why older and smaller firms are more selfish, especially since older firms generally identified as larger and younger firms are generally identified as smaller. Given a low correlation between firm age and firm size, this generalization may be inappropriate.
The manufacturing industry is capital intensive and depends mostly on large batches and
orders. This entails high bills, which induces the use of trade credit. It can therefore be reasoned that the manufacturing industry is familiar with longer trade credit periods. Older firms may therefore become more selfish in my research. This investigation can therefore provide other results when conducted with a sample from other industries than the manufacturing industry. Further research can also focus on other variables that may determine selfish trade credit behaviour.
At last, the year-dummies used seem to indicate a trend between selfish trade credit behaviour
23 References
Altunok, F. (2011). Determinants of trade credit contract terms, Department of Economics, North Carolina State University, North Carolina.
Banerjee, S., Dasgupta, S., and Kim, Y. (2004). Buyer-supplier relationships and trade credit (working paper), Department of Finance, Hong Kong University of Science and Technology, Hong Kong.
Brennan, M.J., Maksimovic, V., and Zechner, J. (1988). Vendor financing. Journal of Finance, 43, 1127-1141.
Cunat, V. (2002). Trade credit: Suppliers as debt collectors and insurance providers. London School of Economics, Financial Markets Group.
Deloof, M. and Jegers, M. (1996). Trade credit, product quality, and intragroup trade: Some European evidence. Financial Management, 25, 33-43.
Emery, G.W. (1984). A pure financial explanation for trade credit. Journal of Financial and
Quantitative Analysis, 19(3), 271-285.
Emmery, G.W. and Nayar, N. (1998). Product quality and payment policy. Review of Finance and
Accounting, 10, 269-284.
Ferris. J.S. (1981). A transaction theory of trade credit use. Quarterly Journal of Financial and
Quantitative Analysis, 22(2), 209-225.
Lee, Y.W. and Stowe, J.D. (1993), Product risk, asymmetric information, and trade credit.
Journal of Financial and Quantitative Analysis, 28, 285-300.
Long, M.S., Malitz, I.B. and Ravid, S.A. (1993). Trade credit, quality guarantees, and product marketability. Financial Management, 22, 117-127.
McMillan, J. and Woodruff, C. (1999). Interfirm relationships and informal credit in Vietnam. The
quarterly Journal of Economics, 114(4), 1285-1320.
Nilsen, J.H. (2002). Trade credit and the bank lending channel. Journal of Money, Credit, and Banking,
34, 226-253.
Ng, C.K., Smith, J.K. and Smith, R.L. (1999). Evidence on the determinants of credit terms used in interfirm trade, Journal of Finance, 54, 1109-1129.
Petersen, M.A. and Rajan, R.G. (1997). Trade credit: Theories and evidence. The Review of Financial
Studies, 10(3), 661-691.
Petersen, M.A. and Rajan, R.G. (1994). The benefits of lending relationships: Evidence from small business data. Journal of Finance, 49, 3-37.
Schwartz, R.A. (1974). An economic model of trade credit. Journal of Financial and Quantitative
Analysis, 9, 643-657.
24
Tewolde, S. and Von Eije, J.H. (2007). Selfishness or co-operation: Trade credit in manufacturing firms in Eritrea. Research in Accounting in Emerging Economies, 7, 369-385.
Wilner, B.S. (2000), The exploitation of relationships in financial distress: The case of trade credit.
25 Tables
Table 1. Descriptive statistics.
The mean, median, maximum, minimum, standard deviation and number of observations are provided for the dependent variable SELFISHNESS in the first column, the main explanatory variable AGE in
the second column and the control variables SIZEt-1, DEBTt-1, NET INCOMEt-1 and RISKt-1 in,
respectively, the third, fourth, fifth and sixth column. In the second column the dependent variable SELFISHNESS is the log of the ratio of accounts payable to total assets plus one minus the log of the ratio of accounts receivable to total assets plus one. The independent variable AGE refers to the age of
a firm, calculated as the panel year - year of incorporation. The control variable SIZEt-1 is the log of
total assets, indicating firm size. DEBTt-1 is the control variable that refers to the ratio of long-term
debt to total assets. NET INCOMEt-1 refers to the control variable calculated as the ratio of net income
to total assets. The control variable RISKt-1 is calculated as the standard deviation of the ratio of net
income to total assets over three years. The subscript t-1 refers the one year lagged-value of the
variables SIZEt-1, DEBTt-1, NET INCOMEt-1 and RISKt-1. The sample consists of all UK
manufacturing firms in Orbis from 2006-2012.
SELFISHNESS AGE SIZEt-1 DEBTt-1
26
Table 2. Possible relations between firm age and selfish trade credit behaviour.
The first two columns represent eight possible interaction between Ln(AP/TA + 1) and Ln(AR/TA+1), representing trade credit behaviour. The dependent variable Ln(AP/TA + 1) is the log of the ratio of accounts payable to total assets plus one. The dependent variable Ln(AR/TA+1) is the log of the ratio of accounts receivable to total assets plus one. The dependent variable SELFISHNESS is the log of the ratio of accounts payable to total assets plus one minus the log of the ratio of accounts receivable to total assets plus one (Ln(AP/TA + 1) − Ln(AR/TA+1)). The interactions between Ln(AP/TA + 1) and Ln(AR/TA+1) in the relations 1-4 indicate selfish trade credit behaviour since accounts payable increases more than accounts receivable increases, or accounts receivable decreases more than accounts payable decreases. The interactions between Ln(AP/TA + 1) and Ln(AR/TA+1) in the relations 5-8 indicate cooperative trade credit behaviour since accounts payable decreases more than accounts receivable decreases, or accounts receivable increases more than accounts payable increases. The third column represents eight possible relations between the interaction of Ln(AP/TA + 1) and Ln(AR/TA+1) and firm age. The independent variable AGE refers to the age of a firm, calculated as the panel year - year of incorporation. A positive relation indicates that the variable SELFISHNESS increases when the variable AGE increases, indicating that older firms are more selfish. A negative relation indicates that the variable SELFISHNESS decreases when the variable AGE increases, indicating that younger firms are more selfish.
SELFISHNESS
Nr. relation Ln(AP/TA + 1) Ln(AR/TA+1) Relation
1. ++5 + Postive 2. + 0 Postive 3. 0 − Postive 4. − −− Postive 5. −− − Negative 6. − 0 Negative 7. 0 + Negative 8. + ++ Negative 5
27 Table 3. Descriptive statistics.
Table 4 . Correlation matrix.
Correlation matrix for the independent variable AGE, the control variables SIZEt-1, DEBTt-1,NET INCOMEt-1 and RISKt-1 and the dummy variables D07, D08,
D09, D10 and D11. The independent variable AGE refers to the age of a firm, calculated as the panel year - year of incorporation. The control variable SIZEt-1
is the log of total assets, indicating firm size. DEBTt-1 is the control variable that refers to the ratio of long-term debt to total assets. NET INCOMEt-1 refers to
the control variable calculated as the ratio of net income to total assets. The control variable RISKt-1 is calculated as the standard deviation of the ratio of net
income to total assets over three years. The subscript t-1 refers the one year lagged-value of the variables SIZEt-1, DEBTt-1, NET INCOMEt-1 and RISKt-1. D07,
D08, D09, D10 and D11 refer to the year-dummies of respectively 2007, 2008, 2009, 2010 and 2011.
AGE SIZEt-1 DEBTt-1
Table 5. The impact of different variables on three regressions determining selfish trade credit behaviour.
In first column the dependent variable Ln(AP/TA + 1) is the log of the ratio of accounts payable to total assets plus one. In the second column the dependent variable Ln(AR/TA+1) is the log of the ratio of accounts receivable to total assets plus one. In the third column the dependent variable SELFISHNESS is the log of the ratio of accounts payable to total assets plus one minus the log of the ratio of accounts receivable to total assets plus one. The independent variable AGE refers to the age of
a firm, calculated as the panel year - year of incorporation. The control variable SIZEt-1 is the log of
total assets, indicating firm size. DEBTt-1 is the control variable that refers to the ratio of long-term
debt to total assets. NET INCOMEt-1 refers to the control variable calculated as the ratio of net income
to total assets. The control variable RISKt-1 is calculated as the standard deviation of the ratio of net
income to total assets over three years. The subscript t-1 refers the one year lagged-value of the
variables SIZEt-1, DEBTt-1, NET INCOMEt-1 and RISKt-1. The sample consists of all UK
manufacturing firms in Orbis from 2006-2012. Each panel regression model controls for firm fixed-effects. The regression includes year dummies (not shown). The numbers in parentheses are p-values.
Ln(AP/TA + 1) Ln(AR/TA+1) SELFISHNESS
30 Table 6. Robustness-test for the model SELFISHNESS.
The SELFISHNESS-CHECK model is the log of the ratio accounts payable to accounts receivable plus one. It provides a robustness-test for the model SELFISHNESS. The independent variable AGE refers to the age of a firm, calculated as the panel year - year of incorporation. The control variable
SIZEt-1 is the log of total assets, indicating firm size. DEBTt-1 is the control variable that refers to the
log of the ratio of long-term debt to total assets. NET INCOMEt-1 refers to control variable calculated
as the log of ratio of net income to total assets. The control variable RISKt-1 is calculated as the
average standard deviation of the ratio net income to total assets over three years. The subscript t-1
refers the one year lagged-value of the variables SIZEt-1, DEBTt-1, NET INCOMEt-1 and RISKt-1. The
sample consists of all UK manufacturing firms in Orbis from 2006-2012. The robustness-test model controls for firm fixed-effects. The regression includes year dummies (not shown). The numbers in parentheses are p-values.
31
Table 7. Relation between SELFISHNESS and year-dummies.
The dependent variable SELFISHNESS is the log of the ratio of accounts payable to total assets plus one minus the log of the ratio of accounts receivable to total assets plus one. The year-dummies D07, D08, D09, D10 and D11 represent, respectively, year-dummies of 2007, 2008, 2009, 2010 and 2011.
32 Appendixes
Appendix 1. Example using bank loan to profit from providing trade credit.
Customers 1, 2, 3 and 4 are fictitious customers that are all customers of a firm selling goods. The
value of goods indicates the amount of money a customer pays when paying with cash immediately.
The value goods credit refers to the amount of money a customer pays when using trade credit. The value of goods is increased with the implicit rate for the length of the credit period. In this example, a credit period of one month is selected. The implicit rate period means the implicit rate paid over the value of goods for the length of the credit period. The yearly implicit rate is the rate paid over the value of goods when the credit period would entail one year instead of one month. All implicit rates are rounded to two decimals. Furthermore, the firm selling goods provides trade credit by using bank credit. This bank loan has a value of € 10,000.00, similar to the total value of goods provided to the customers. The market rate for the bank loan is 5% per year, 0.41% for a credit period of one month. Customer 1, 2 and 3 are charged with implicit rates above the market rate. This entails that the firm selling goods on credit, profits from providing trade credit with the use of a bank loan. Customer 4 has an implicit rate below the market rate. This entails that the firm selling goods on credit, loses 0.02% in one month, when providing trade credit to customer 4 with a bank loan. Customer 4 may be a customer that would otherwise not participate in trade without a low interest rate. Now the firm will benefit from increased sales by losing 0.02% as the difference between the one month implicit rate and the one month market rate. On overall, the firm selling goods on credit profits from providing trade credit through a bank loan since the average implicit rate is 0.44% for one month, 5.38% for one year, which is more than the market rate of 5%. This indicates that the firm profits 0.03% (0.44% - 0.41%) in one month and profits 0.38% on a yearly basis (5.38% - 5%).
Implicit rates are absorbed in the price of the goods. The credit providing firm anticipates that
33
customer 1 customer 2 customer 3 customer 4 Total
Value of goods € 2,500.00 € 3,000.00 € 2,000.00 € 2,500.00 € 10,000.00
Value goods credit € 2,511.18 € 3,012.70 € 2,009.74 € 2,509.79 € 10,043.40
Implicit rate period 0.45% 0.42% 0.49% 0.39%
Yearly implicit rate 5.50% 5.20% 6.00% 4.80%
Firms also provide trade credit options or discounts. These discounts are used to trigger customer to pay earlier. In this way, money can be reinvested for other investment opportunities. Bills indicate for example: 1/10 net 30. This means that the bill has to be paid within 30 days (net 30) but a discount of 1% is received when the customer pays the bill within 10 days (1/10). For example, customer 1 pays within 10 days instead of 30 days. Then this customer has to pay €2,486.07 (0.99* € 2,511.18). instead of €2,511.18. Rather than paying an implicit rate of 0.45%, customer 1 now receives an interest rate of 0.55% (0.45% − 1% = -0.55%) from the firm selling goods. This is called a negative interest rate, which is 22% on a yearly basis. To prevent implicit rates to become negative when providing discounts, the supplier thus has to charge very high initial implicit rates.
Calculating the price for information signals by using discounts requires to take into regard
34 Appendix 2. Descriptions abbreviations and concepts.
AP Accounts payable AR Accounts receivable TA Total assets LTD Long-term debt NI Net income i Firm t Panel year
YINCi Year of incorporation
SELFISHNESS Selfish trade credit behaviour
AGE Firm age
SIZEt-1 Firm size
DEBTt-1 Firm's long-term debt
NET INCOMEt-1 Firm's net income
RISKt-1 Firm's net income uncertainty
35
Appendix 3. Formulas used for dependent and independent variables.
SELFISHNESSit Ln((AP/TA)it + 1) - ln((AR/TA))it + 1)
AGEit t - YINCi
SIZEi,t-1 Ln(TA)i,t-1
DEBTi,t-1 (LTD/TA)i,t-1
NET INCOMEi,t-1 (NI/TA)i,t-1
RISKi,t-1 σ3(NI/TA)i,t-1
