Do the Better Financing Abilities of Listed Firms affect Trade Credit?

(1)

Do the Better Financing Abilities of Listed Firms affect Trade Credit?

M.F.J. Kalff

1

Student number: 1717391

Master thesis MSc BA Finance

Faculty of Economics and Business, University of Groningen

The Netherlands, June 2012

Abstract

This study explores whether there is a significant difference in the ratio of accounts receivable and accounts payable to total assets between listed and unlisted firms. More specifically, it tests the hypothesis that listed firms provide more trade credit and use less trade credit compared to unlisted firms as a result of their access to non-bank funding. When controlling for other determinants, listed firms do indeed provide more trade credit and use less trade credit than unlisted firms. Furthermore evidence is found that the financing ability effect is more prevalent in market driven countries compared to bank driven countries.

JEL Classifications: G32, M21

Keywords: Trade credit, Accounts payable, Accounts receivable, Financing ability effect

Supervisor: Dr. J.H. Von Eije

1

(2)

2 INTRODUCTION

Trade credit, measured by accounts receivable and accounts payable in the balance sheet of a

firm, is based on agreements between sellers and buyers that allow buyers to receive goods or services without making an immediate payment. It separates the exchange of goods and their payments in time. Trade credit varies substantially across firms and industries and substantial research exists that tries to clarify this variation (Mian and Smith, 1992; Rajan and Zingales, 1995; Petersen and Rajan, 1997; Burkart and Ellingsen, 2004; Bougheas et al. 2009).

Receiving trade credit can be a good alternative for firms who are unable to raise capital through traditional channels such as bank loans (Petersen and Rajan, 1997). This can be true for relatively small firms but also for unlisted firms compared to listed firms, and firms all over the world consider this as an important element of their corporate financing (Bougheas et al., 2009). In the early 1990’s Rajan and Zingales (1995) reported that the combined trade credit of US firms was 17.8% of total assets. Also for other countries such as Japan, Germany, Italy, UK and France trade credit is a major part of their assets (respectively 22.5%, 26.9%, 29%, 22.1% and 28.9% (Rajan and Zingales (1995)). Trade credit has already been investigated from many perspectives and profitability, size, industry, country, leverage, interest rates and growth are variables that have been used to explain the different levels of trade credit (Mian and Smith, 1992; Rajan and Zingales, 1995; Petersen and Rajan, 1997; Burkart and Ellingsen, 2004; Bougheas et al. 2009).

Until now literature focused in particular on firms that are listed on a stock exchange. In this research the main focus is not on variables that influence or are affected by the use and providing of trade credit, but I explicitly look at the difference between listed and unlisted firms. The aim of this paper is to compare the ratio of payables and receivables between listed and unlisted firms and test the

hypothesis that the relatively easy access to non-bank financing of listed firms influences the amount of trade credit received or provided. I test the ‘financing ability hypothesis’ against the ‘financing needs hypothesis’. The financing needs hypothesis suggests that listed firms have significantly less receivables and more payables compared to unlisted firms. This would be caused by the fact that firms with high growth opportunities and investments are induced to go public (Pagano, Panetta and

(3)

3 would take place. First, companies that have gone public can overcome borrowing constraints; they now have an improved access to alternative sources of financing other than banks. And second, they have greater bargaining power with banks. By gaining access to the stock market they create

competition amongst types of financing other than bank debt and thereby ensuring a lower cost of credit.

Almeida et al. (2004) found evidence that financially unconstrained firms have no need to maintain large cash holdings for future investments and debt obligations, and it implies that corporate liquidity becomes less relevant.

When comparing the use and extension of trade credit between firms, literature written so far focused on small versus larges firms. In this research the main objective is to test whether the financing ability hypothesis hold for listed in comparison to unlisted firms. To test this I apply several tests. This study first tests whether there is a significant difference for the receivables ratio between listed and unlisted firms. The same test is repeated for the payables ratio between listed and unlisted firms. The second test I apply is a country comparison, to test whether there are differences in ratios between countries. Subsequently, a distinction is made between countries that are driven by the market and countries with active banking systems. In order to get the most accurate results, these tests are made on a global scale consisting of 2283 unlisted firms and 2910 listed firms, resulting in a total of 5193 firms divided over 23 countries. Only large firms are included because listed firms are on average larger than unlisted firms. Furthermore the sample consists only of manufacturing firms as these types of firms make the most use of trade credit. I control for firm characteristic such as profitability, economies of scale, cash holdings, leverage, investment rate and the riskiness of the cash flows, next to that I include year, country and industry dummies.

The results show that the payables ratio for listed firms is significant lower than for unlisted firms. This is what I would expect when the financing ability effect is present. In order for the financing ability effect to be present the receivables ratio has to be higher for listed than for unlisted firms. Results show that the receivables ratio for listed firms is indeed higher for listed firms than for unlisted firms. These two results combined confirm that the financing ability effect is present, where listed firms use less trade credit and provide more trade credit and this is likely caused by their easy access to alternative sources of financing.

This paper is structured as follows. Section 1 describes the literature written so far concerning the cost and benefits of trade credit and discusses several theories on why firms would provide trade credit. Section 2 focusses on the financing ability effect and why there would be any differences regarding the receivables and payables ratio between listed and unlisted firms across countries. In section 3 the hypotheses are explained. The data description and the methodology will be explained in section 4. In section 5 you will find the results of the multiple regressions and the country comparison and,

(4)

4 1. TRADE CREDIT

Cash inflows are delayed when firms provide trade credit to their customers. So why would firms provide credit in the first place? Several authors have come up with theories that try to explain the existence of providing trade credit. The next section will first describe these theories, and after that the costs of providing trade credit and the benefits and costs of using trade credit will be discussed.

1.1 Theories and benefits of providing trade credit 1.1.1 Financing motive

Schwartz (1974) proposed in 1974 a financing motive for the existence and use of trade credit. The seller may have an advantage over traditional lenders (such as banks) in examining the credit worthiness of his buyers and as well the ability to superior monitoring and superior ability to force repayment of the credit. The theory suggests that because of the financing motive, trade credit will mostly flow from firms that have relatively easy access to capital markets to firms that are restricted in their ability to obtain additional funds. This financing advantage comes from three sources.

Sellers may have better information on their buyers than financial institutions have on the same buyers. They gain extra information about the financial strength of the company as a by-product that comes with the selling. Second, a seller often has a network of the same sort of buyers from which he can get additional information, and in that case the seller may also be better in assessing the financial strength of a buyer than a financial institution. And thirdly, a supplier has the power to cut off further supply when the repayments by the buyer are stopped. This is especially effective when the supplier is one of the few suppliers in the industry. When cutting off future supply the buyer will most of the time not be able to continue daily operations. Financial institutions are also able to withdraw further

payments but this will not have an immediate effect on the buyer’s operations (Petersen and Rajan, 1997). Furthermore, if a buyer can’t make its payments anymore to the supplier, the supplier can seize the goods that were supplied. In comparison, the collateral is more valuable for a manufacturer (supplier) than for an institutional lender because it’s easier for the manufacturer to sell the collateral to a third party than for a financial institution. (Mian and Smith, 1992)

1.1.2 Transactions cost theory

In a world with perfect certainty and without any transactions costs this theory would not be relevant, but in the current environment firms do have to deal with uncertainty about the future and transaction costs (Ferris, 1981). Providing trade credit may reduce these transaction costs. As goods flow

(5)

5 payments to periodic bills is the better foresight a firm will have on the availability of cash holdings. This will cause the predictability of the cash in- and outflows to increase.

1.1.3 Inventory costs

From another transactions cost perspective, providing trade credit also reduces inventory costs. In the current environment it is hard to predict future demand, and firms face variable demand over time (Emery, 1987). This is especially true when there are strong seasonalities in the firm’s products. In order to maintain an even production cycle the supplier has to build up bulky inventories. This

inventory results in two costs, costs for additional capacity and costs to finance it (Petersen and Rajan, 1997). A way for a firm to deal with these inventory costs is to provide trade credit. Providing trade credit can encourage buyers to buy goods in off-season periods. The buyer will receive a discount if the credit is repaid within the certain “discount period”. So the supplier faces a tradeoff between inventory cost and offering trade credit (Bougheas et al. 2009). According to this trade-off theory providing and receiving trade credit can be beneficial for both the supplier and the buyer. As explained above, the supplier will save on additional inventory costs that would have to be made when he would not have provided trade credit and the buyer will save money through the received price reduction (only when he pays within the discount period).

1.1.4 Price discrimination

Even if suppliers do not have a financing advantage over financial institution they can still use trade credit through effective price discrimination (Schwartz and Whitcomb, 1979; Brennan, Maksimovic, and Zechner, 1988; Mian and Smith, 1992). These authors assume that the predetermined credit terms are indifferent to the creditworthiness of the buyer (borrower). Since trade credit exposes the seller to the risk that the buyer will not be able to repay the credit (default risk), it reduces the effective price for buyers with a low level of creditworthiness. And thus, in a price elastic market, providing trade credit will result in an advantage for the seller. So, a seller is able to attract customers that would not be attracted without offering trade credit at low interest rates. Petersen and Rajan (1997) find evidence that this theory is especially true for high profit-margin firms. Firms with high profit margins are able to offer trade credit at low interest rates, and thus with a greater loss on their credit than financial institution can.

1.1.5 Enhancing relationships

(6)

6 to avoid risk of being seen as a bad trade partner (McMillan and Woodruff, 1999; Tewolde and von Eije, 2006)

The theories described above try to explain the existence of trade credit. Better information than financial institutions, effective price discrimination, enhanced relationships and reduced transaction and inventory costs are the benefits of providing trade credit (Schwartz, 1974; Ferris, 1981; Emery, 1987; Brennan, Maksimovic, and Zechner, 1988; Mian and Smith, 1992; Petersen and Rajan, 1997; McMillan and Woodruff, 1999; Tewolde and von Eije, 2006). Many theories are developed to give reasons why firms provide trade credit. Much less is written about the costs of providing trade credit and the benefits and costs of using trade credit. The following section discusses the costs of providing credit and the benefits and cost of using trade credit.

1.2 Costs of providing trade credit

Nadiri (1969) mentioned three sorts of costs for sellers that provide trade credit, namely carrying costs, deprecation costs and the credit standards of the lender. The carrying cost is the forgone real income that is tied up in accounts receivable. The deprecation costs consist out of two parts. First a certain fraction of the total amount of accounts receivable has to be written off the books due to bad debts. Secondly, substantial costs have to be made for buyers who do not pay within the specified payment period. Lastly, sellers must finance their receivables without jeopardizing their own liquidity position. When sellers provide trade credit to buyers, sellers will not have any cash inflows from these sales but instead, do have to pay the costs of producing the goods. These cash outflows without any cash inflows will cause opportunity costs, since the provided trade credit could have been used for investment opportunities.

Providing trade credit will cause sellers also to have some additional administrative costs (Biais and Gollier, 1997). Sellers must allocate energy, time and resources for assessing the credit worthiness of buyers and structuring delayed payment contracts. After the contract has been set up sellers must also devote some time in the collection process (Mian and Smith, 1992, 1994). Furthermore, when a firm has much debt obligations, the delayed cash inflows could have been used to pay off debt or interest.

1.3 Costs and benefits of using trade credit

(7)

7 often more expensive ways of financing. Combined with minimizing receivables and maximizing payables this policy causes cash inflows to accelerate and cash outflows to be delayed. In this case trade credit can be used as a cheap way of financing (Tewolde and Von Eije, 2006).

As with accounts receivable, accounts payable also has its marginal costs. Using trade credit from your supplier often means that you accept higher interest rates over your received credit. Burkart and Ellingsen (2004) found that many firms repeatedly fail to benefit from early payment discounts given by the supplier and eventually end up with borrowing from their supplier at annual interest rates exceeding 40%. Research from Jain (2001) and Petersen and Rajan (1997) shows that firms that make use of trade credit pay higher interest rates on trade credit than on direct loans from banks. There is also the risk that suppliers do not want to do business anymore with buyers because payments are repeatedly not paid on time, and eventually this will lead to damaged relationships. Contrary, when buyers do pay on time, they will not only benefit from the cash discounts that firms get when they pay within the discount period but they will also enhance their relationship with the suppliers.

2. FINANCING ABILITY EFFECT

(8)

8 according to the financing ability effect. Managers from unlisted firms who maximize firm value chose the optimal level of accounts receivable or payable represented by point A*.

Figure 1. Financing ability effect accounts receivable

Figure 2. Financing ability effect accounts payable

(9)

9 For accounts payable the financing ability effect is shown in figure 2 by a shift to the left. The amount of payables will shift from curve AA, which again represents the unlisted firms, to curve BB. Meaning that listed firms will have less payables than unlisted firms as a result of easy access to alternative sources of funding other than credit provided by banks. In this case managers who maximize firm value chose the optimal level of accounts payable B*, where receiving discounts on early payments and building relationships with suppliers are the main benefits of low accounts payable.

2.1 Market driven -versus bank driven countries

Now that the financing ability effect is defined I will explain the differences that one can expect between countries. If I find evidence for the financing ability effect it is of interest whether this effect takes place in all of the countries, just a few or only in countries with the same characteristics. To test this, I categorize the countries into market driven and bank driven countries. I will use the

classification that is made by Demirguc-Kunt and Levine (1999). They used the following procedure. In order to classify countries into market driven and bank driven they analyzed size, level of activity and the efficiency of financial systems and how they differ across countries. Secondly, they defined different indicators and observed divers patterns of financial structure across countries. And thirdly, the legal, regulatory and policy determinants of financial structure were investigated. Their results show that countries with a Common Law tradition, strong protection of shareholders, good accounting standards and low levels of corruption tend to be market-driven countries. Furthermore they found that in market driven countries stock markets and banks collaborate in getting the public savings to firms, utilizing corporate control and enabling risk management. Contrary, countries with Civil Law traditions and poor protection of shareholders tend to be bank-driven countries. In bank driven countries the leading role is given to the banks. Banks mobilize savings, allocate capital, oversee investment decisions and provide risk management vehicles.

The question is now, when I find evidence of the financing ability effect, if this effect is significantly more present in market driven countries, bank driven countries or that there is no significant difference between listed and unlisted firms within those two different kinds of countries.

(10)

10 3. HYPOTHESES

In order to test whether the financing ability effect is indeed present between listed and unlisted firms two null hypotheses are formulated.

HA0: The receivable ratio will be equal between listed and unlisted firms.

If the null hypothesis is rejected, two different results are possible. Receivables are higher in listed firms than in unlisted firms, which would mean that the first part regarding the financing ability effect would be confirmed. Results could also show that listed firms have a lower receivables ratio compared to unlisted firms, which would mean that we should dismiss the presence of the financing ability theory. When H0 can’t be rejected this will mean that listed and unlisted firms follow the same sort of trade credit policy, or that it simply doesn’t matter whether a firm is listed or not.

In order to test whether the financing ability effect is present, I want to know if the receivables ratio is significantly higher for listed firms than for unlisted firms. Therefore the following alternative

hypothesis is formulated:

HA1: The receivables ratio in listed firms is higher than in unlisted firms.

When this hypothesis cannot be rejected I can conclude that the financing ability effect for the receivables is present. When it is rejected, I conclude that the financing ability effect is not present on the receivables side and that other factors may have a larger impact on the decision how much trade credit to provide.

I repeat the same test and hypothesis for the payables ratio, which results in the following null hypothesis:

HB0: The payables ratio will be equal between listed and unlisted firms.

The same counts for this hypothesis, when the null hypothesis can be rejected this will either mean that payables are higher in listed firms or that payables are lower in listed firms. In the latter case I would assume that the financing ability effect for the payables is present. In order to test whether the payables ratio is smaller for listed firms than for unlisted firms I use the following alternative hypothesis:

HB1: payables ratio in listed firms is smaller than in unlisted firms.

(11)

11 Lastly, I will make a comparison between market-driven countries and bank-driven countries. I will not only test in which countries the financing ability effect is present but also whether there is a significant difference of using and providing trade credit between listed firms in market driven countries and bank driven countries.

To test whether listed firms in market driven countries provide more trade credit and use less trade credit compared with listed firms in bank driven countries I test the following null hypothesis.

HC0: The receivables ratio will be equal between listed firms located in market driven countries and listed firms located in bank driven countries.

When the null hypothesis cannot be rejected this will mean that firm policies regarding accounts receivable are not influenced by country characteristics. If the null hypothesis is rejected, the receivables ratio can be either higher or lower in market driven countries. In order to rest whether listed firms in market driven countries provide more trade credit than listed firms in bank driven countries, the following alternative hypothesis is used:

HC1: The receivables ratio is significantly higher for listed firms located in market driven countries compared to listed firms located in bank driven countries.

When this hypothesis cannot be rejected, I can conclude that listed firms in market driven countries are able to provide more trade credit than listed firms in bank driven countries, caused by a more active stock market where firms have better access to alternative sources of funding.

I repeat the same test and hypothesis for the payables ratio, which results in the following null hypothesis:

HD0: The payables ratio will be equal between listed firms located in market driven countries and listed firms located in bank driven countries.

Again, when the null hypothesis cannot be rejected this will mean that firm policies regarding

accounts payable are not influenced by country characteristics. If the null hypothesis is rejected, firms listed in market driven countries can either use more or less trade credit than firms listed in bank driven countries. To test this I use the following alternative hypothesis:

HD1: The payables ratio is significantly smaller for listed firms located in market driven countries compared to listed firms located in bank driven countries.

(12)

12 4. DATA & METHODOLOGY

The data for this research is extracted from the Orbis database of Bureau van Dijk. This database has been used because it contains financial information of listed companies as well as unlisted companies. The data had to meet several criteria. First of all, all the values for listed and unlisted companies must be available for total assets, debtors (accounts receivable), creditor (accounts payable), cash and cash equivalents and operating revenue (turnover) for the period 2005 until 2010. For a better comparison between listed and unlisted firms I also required that all firms have a value for total assets above 100 million euro. This requirement is made since listed firms are on average much larger than unlisted firms and in order to make a good comparison all firms are required to have a large amount of assets. Only manufacturing firms are chosen as these firms make the most use of trade credit. Firms are included with NACE rev. code 2 industry classifications 20 until 39. In this way I exclude financial firms and utilities, agricultural, building and service firms. After examining the data I also excluded all companies with industry classifications 21 and 35, which means that pharmaceutical and

electricity/gas/steam conditioning supply firms are also excluded. In order to compare differences between countries, only those countries are selected that have at least ten listed and ten unlisted firms headquartered in that country. This selection procedure produced 5176 firms from which 2906 are listed and 2270 are unlisted, divided over 23 countries2 and six main industries3. The total number of firm-year observations for which the dependent variable is available is 31158, from which are 17460 listed firms and 13698 unlisted firms.

Table 1 provides an overview of the number of firms by type of firm and by country and the mean receivables and payables ratio for both types of firms (listed and unlisted). Also the p-values of a test on equality by classification are given.

Table 1 about here

Most manufacturing firms that are part of this dataset are located in Japan (1150), the United States (603) and China (414), for European countries most manufacturing firms can be found in Italy (378), France (337) and Germany (323). Listed firms have on average a receivables ratio of 17.1%, while unlisted firms have a receivables ratio of 19.4%. Total assets of listed firms consist for 11.5% of payables and 14.6% for unlisted firms. Meaning that listed firms have on average 2.3% less

2_{Austria, Belgium, Brazil, China, Germany, Spain, Finland, France, Great-Britain (UK), Greece, Italy, Japan,}

Korea, The Netherlands, Norway, Poland, Russia, Sweden, Singapore, Thailand, Taiwan, Ukraine, United States.

3

(13)

13 receivables than unlisted firms, and have on average 3.1% less payables compared to unlisted firms. The p-value indicates whether there is a significant difference between listed and unlisted firms, calculated for both ratios. In 17 of the 23 countries the receivables ratio for listed and unlisted firms differs significantly from zero at a 5% level. For the payables ratio between listed and unlisted firms 13 of the 23 countries differ significantly. In China I found the lowest average receivables ratio for listed firms (9.7%), while firms that have the highest average receivables ratio for listed firms are located in Japan (23%). The lowest and highest average receivables ratio for unlisted firms is found in respectively the United States (6.6%) and France (26.3%). Listed firms in Brazil have on average the lowest average payables ratio (7.6%) with Italy having the highest payables ratio (17.2%). Unlisted firms in the United States have on average the lowest payables ratios (3.8%) and in China the highest (26.4%).

4.1 Control variables

Until now literature has shown that there are several factors that influence firms’ decisions about how much trade credit to use and to provide. Therefore it is important to include these factors as control variables. The dependent variables used in this research are the receivable ratio and the payables ratio. Both variables are calculated by dividing both accounts (payable and receivable) by the total assets of the firm.

According to NG, Smith and Smith (1999) the size of a firm has significant impact on the decision to provide or not provide trade credit to particular costumers due to economies of scale. Investigating the creditworthiness of customers and managing outstanding credit includes fixed costs. This means that large firms with relatively many customers are able to spread these fixed costs over much more

customers than firms that do not have large a customer’s base. Additionally, the more customers a firm has, the more information the firm is able to acquire over the related industry and the default risk of the customers’ rivals. They found evidence that size has a positive relation with accounts receivable. Banerjee et al. (2004) found that the size of a company influences accounts receivable as well as accounts payables. They found that larger firms tend to provide more trade credit and use less trade credit in comparison with smaller firms. In addition, Peterson and Rajan (1997) found evidence that small firms do not only use less trade credit, they also provide less credit. Furthermore they found a weak positive relationship between firm size and the offering en usage of trade credit. Because several studies found positive relations I include size, measured by the natural log of sales, as a control variable.

(14)

14 Preve (2009) found that firms facing profitability problems increase accounts payables, but when they face cash flow problems they decrease receivables and thereby speeding up the collection process. Actually I would expect the opposite, firms with higher profits are more able to provide credit, and thus have larger accounts receivables. Deloof (2003) found a negative relation between profitability and accounts payable, consistent with the view that firms with less profit will have higher levels of accounts payable. Because there is still some discussion which relationship profit has with accounts receivable and payable, a profit ratio is included as a control variable. This ratio is measured as net income divided by the total assets of the firm.

Leverage should also have an impact on the decision how much credit to provide and use. According to Cunningham (2004), firms use trade credit when their credit lines become exhausted. This would mean that when firms have large debt ratio they would use trade credit as substitute for bank credit. From this reasoning I expect that when leverage increases, accounts payable will also increase. For accounts receivable I would expect to find the contrary. When leverage increases, firms are less inclined to provide trade credit. As leverage increases the firm also has higher repayment obligations, in order to repay these obligations they will prefer their customers to pay by cash instead of providing them credit. It’s also possible that we would see contrary results when a firm is in financial distress. In that case, a firm could also provide more trade credit despite of having a large debt ratio and trade credit could then be used to increase sales. Because I expect leverage has a substantial influence on accounts receivable and payable I include leverage as a control variable, measured by dividing the firms’ long term outstanding debt by total assets.

I also include a control variable that measures the sensitivity of the cash flow. When cash flows of firms become more volatile firms become more risky. In order for firms to avoid financial distress cost it would be wise to collect payments as quick as possible and thus require their customers to pay directly. A negative relation is expected between the riskiness of cash flow and accounts receivable. The contrary result I expect to find with accounts payable. When cash flows are risky, firms are likely to delay payments as long as possible. Therefore a positive relation is expected between the riskiness of cash flows and accounts payable. Because the database I use has too many missing values for calculating the cash flow of a firm, net income is used as a proxy for cash flow. The riskiness of cash flow is measured by standard deviation of net income of the preceding year and the current year. This also means that each first year observation will fall out, though because of the large sample this is not likely to influence the results.

(15)

15 of capital expenditures for unlisted firms. Therefore I use the change in fixed assets of the firm. The investment rate of the current year is measured by the amount of fixed assets of next year minus the amount of fixed assets of the current year divided by the amount of fixed assets of the current year. This means that the each last year observation will be lost, as with the riskiness of the cash flow, this will not influence the results because of the large sample.

Furthermore, cash and cash equivalents are included as control variable. Firms that are increasing their receivables delay their cash inflows, which results in less cash, therefore I expect a negative relation between cash and accounts receivable. Opposing effects are expected with accounts payable. Increasing payables will lead to delayed cash outflows, which results in an increase of the cash holdings of a firm. The cash ratio is measured by cash and cash equivalents divided by total assets. At last I also include the payables ratio and the receivables ratio in the regression for measuring the receivables ratio and payables ratio respectively. Banerjee et al. (2004) found evidence that firms match the structure of their assets and liabilities by allowing the payables policy be determined partially by their receivables. I therefore expect both ratio’s to have a positive relationship with each other.

4.2 Methodology

To test the effect that the determinants have on the payables and receivables ratio of a firm and

whether the financing ability effect is present, I estimate the following regression using Ordinary Least Square (OLS). The equations for both dependent variables are as follows:

For the receivables to assets ratio:

(1)

The same regression is used for the payables to total assets ratio, only payable and receivable ratio switch place:

(2)

(16)

16 ί. NITA is net income divided by total assets for firm ί. CATA is the cash ratio, calculated by dividing cash and cash equivalents by total assets for firm ί. DEBT is measured by total long term debt divided by total assets for firm ί. INVST is the investment rate of the firm, calculated by the relative change of future fixed assets for firm ί. RISK is the riskiness of the cash flows, measured by taking the standard deviation of net income of the previous and current year for firm ί. DYEAR, DCOUNTRY and DINDUSTRY are dummy variables for respectively years (6), countries (23) and industries (6). is the error term of the equation.

In the second part of this paper I compare market driven countries with bank driven countries. In order to test the difference of the receivables and payables ratio for listed between market and bank driven countries, the following hypotheses are used.

For the receivable ratio:

(3)

For the payables ratio:

(4)

MARKET is a dummy variable that takes the value 1 if firm ί is located in a market driven country and 0 otherwise.

4.3 Descriptive statistics

In table 2 the descriptive statistics for the dependent and independent variables are listed. It shows the number of observations, the mean, standard deviation, median and the minimum and maximum.

Table 2 about here

(17)

17 For the individual samples, listed firms have a mean receivables ratio of 17.1%, which is significantly lower than the mean receivables ratio of unlisted firms (19.6%). The mean payables ratio for the listed firms is 11.5% which is also significantly lower than the ratio for unlisted firms (14.6%). According to the financing ability effect listed firms have more receivables and less payables compared to unlisted firms. Table 2 indicates that listed firms do indeed have less payables than unlisted firms, but also shows that listed firms have less receivables compared to unlisted firms. From these results I can conclude that without the inclusion of control variables and dummies the financing ability effect does not exist for listed firms. So without the use of any control variables I have to reject the alternative HA1 hypothesis which states that listed firms have a higher receivables ratio than unlisted firms. To check these results I also performed a test based on an OLS regression corrected for autocorrelation and heteroskedasticity with the Newey-West estimator. It shows that listed firms have a receivable ratio that is significantly lower (-0.026) at a 1% level than the receivable ratio for unlisted firms. Also the payables ratio is significantly lower (-0.031) at a 1% level for listed firms than for unlisted. These results are consistent with the results from table 2. The regressions can be found in the appendix (table 3). Furthermore, the statistics show that listed firms are significantly larger than unlisted firms, both for the mean, which is tested with an equality t-test, and for the median, tested with a Mann-Whitney Wilcoxon test. The mean net income to total asset for listed firms is smaller in comparison with unlisted firms, although the median for that ratio is higher for listed firms. Despite the winsoring, this difference is probably caused by outliners, as unlisted firms have a maximum of 56.3% of total assets and listed firms only 24.3%. Listed firms also have significantly larger cash holdings than unlisted firms, showing an average of 14.2% for listed firms and 6.8% for unlisted, this finding is line with Von Eije (2012). Listed firms have a slightly higher mean for long term debt to total assets ratio, but a median that is 6.9% higher. Measured by the relative change in fixed assets, listed firms significantly invest on average 2.8% more than unlisted firms. Lastly, the mean and median for risk, measured by the standard deviation of net income is lower for listed firms.

Table 4 shows the correlation between the independent variables for both listed and unlisted firms.

Table 4 about here

(18)

18 payables and accounts receivable. This effect is much less observable for listed firms.

4.4 Development of variables over time

Table 5 shows the development of the variables for listed and unlisted firms over the period 2005 until 2010. When observing these variables over time there is one notable development that influences almost all variables. Around 2007 and 2008 the receivables ratio, net income ratio and the investment rate of listed and unlisted firms decrease significantly. In the same period the payables ratio and the riskiness of listed and unlisted firms increase significantly. This has most likely to do with the credit crises that started in the United States in the summer of 2007 and found its way quickly after that into Europe. Although the largest effects were on financial institutions which are excluded from this sample, manufacturing firms were also negatively influenced as a result of the credit crises.

Table 5 about here

(19)

19 5. RESULTS

Table 6 shows the regression results for the receivables ratio of the entire sample where both listed and unlisted firms are included. The standard errors of the coefficients are estimated using the Newey-West estimator. Dummies are also included for 22 countries, five industries and five years.

Table 6 about here

The payables and receivables ratios for listed firms, represented by LISTED, have changed compared to the ratios that were observed from the descriptive statistics. The descriptive statistics in table 2 showed us that the receivables ratio for listed and unlisted firms were respectively 0.171 and 0.196, where listed firms have on average 2.5% less receivables, which did not confirm the financing ability effect. Strikingly, when we include all control variables and dummies, results show that listed firms do in fact have on average a higher receivables ratio than unlisted firms. The coefficient for listed firms has changed from -0.025 in the univariate regression to 0.01 in the multiple regression test, and at a significance level of 1%. The coefficient for the payables ratio still has a negative sign though the effect has become somewhat stronger. The coefficient changed from - 0.031 to -0.036. This result shows us that listed firms have on average a significant lower payable to total assets ratio. So, when controlling for size, net income, cash & cash equivalents, long term debt, investment rate and riskiness of the cash flows and include dummies for countries, industries and year, the financing ability effect is confirmed by the regression results: the receivables ratio is significantly higher for listed firms and the payables ratio is significantly lower for listed firms compared to unlisted firms. I can conclude that hypotheses HA1 and HB1 are accepted.

The results also tell us something about the way the determinants influence the payables and receivables ratio of listed and unlisted firms pooled together. Banerjee et al. (2004) found evidence that firms match the structure of their assets and liabilities by allowing the payables policy be

determined partially by their receivables and the other way around. This is confirmed in table 6, with a coefficient of 0.58 for the relation between the payables ratio and receivables ratio, and a coefficient of 0.407 for the relation between receivables and payables both significant at a 1% level.

Size has a significant negative relation with the receivables ratio, meaning that when firms become larger they provide less trade credit. This contradicts the results of Peterson and Rajan (1997), NG, Smith and Smith (1999), Banerjee et al. (2004) and who found a positive relation, although their evidence is based on firms located in the US, which likely influences the results. The same

(20)

20 Net income shows a significant positive relation with receivables. This confirms the results of Deloof (2003) and Bougheas et al. (2008) who stated that firms with high profit tend to provide more trade credit. The net income ratio indicates a negative relation with payables which is also in line with the findings of Deloof (2003) who stated that firms with high profits use less credit. Though these result contradict the finding of Bougheas et al. (2008) who found that firms with high profits would use more trade credit.

The cash ratio has a negative coefficient for both the receivables and the payables ratios although the relation between cash and payables is insignificant. I expected to find the opposite result, when cash holdings increase, firms are more willing to provide credit. There is a possibility that this result is caused by reversed causality. As firms provide more trade credit, cash flows are delayed, which causes cash holdings to decrease. The results also indicate that as the cash holdings of firms become larger, firms make less use of trade credit, confirming the predictions of Bougheas et al. (2008).

The negative relation between the level of debt and the receivables ratio confirms the findings of Cunningham (2004), meaning that when firm have higher levels of debt they tend to provide less trade credit, thereby speeding up their cash inflows in order to fulfill their debt obligations .The level of debt also has a significant negative relation with payables. These results contradict the results of

Cunningham (2004) who found that firms with higher levels of debt will use trade credit as a substitute for their exhausted credit lines.

When the investment rate increases, so does the receivables ratio at a 1% significance level. According to Molina and Preve (2009) and Peterson and Rajan (1997) firms can increase sales by providing more trade credit. Revenues coming from these additional sales can be used for the planned investments. On the other hand, investments have a positive relation with payables. Firms need funds in order to invest, by making more use of trade credit they delay their cash outflows which subsequently can be used for the investment.

The exposure to risk of a firm is negatively related to the receivables ratio. When cash flows become uncertain firms try to speed up the collection process by providing less credit, and thereby reducing the receivables ratio. This is in line with findings of Bougheas et al. (2008). On the other hand, risk is positively related to the payables ratios, meaning that when cash flows become uncertain, firms make more use of provided trade credit in order to delay their payments to suppliers and thereby decreasing the riskiness of their cash flows. This is contrast with the findings of Bougheas et al. (2008) who found a positive relation between risk and accounts payable.

5.1 Market-driven and bank-driven countries

(21)

21 tradition, strong protection of shareholders, good accounting standards and low levels of corruption tend to be market-driven countries. Contrary, countries with Civil Law traditions and poor protection of shareholders tend to be bank-driven countries. Table 7 shows the ratios of receivables to assets and payables to assets, after controlling for size, profit, cash and cash equivalents, debt and investment rate, and including dummy variables for industries and years. A distinction is made for market-driven countries4 and bank-driven countries.

Table 7 about here

In this sample there are 23 countries from which are 9 market-driven countries and 14 bank-driven. In 10 countries5 listed firms have significantly more receivables than unlisted firm, and in 15 countries6 listed firms have significantly less payables than unlisted firms. When comparing these results, table 7 shows that in 12 of the 23 countries the financing ability effect, where listed firms have significantly more receivables and less payables than unlisted firms, is present. These countries are Austria, Belgium, Brazil, Germany, Finland, United Kingdom, Japan, Korea, Poland, Sweden, Thailand and the United States. For Brazil, Germany and Sweden the payables ratio is not significant but does show the right coefficient. From these 12 countries, six of them are market-driven countries. This means that in 66.67 % (6 out of 9) of the market-driven countries the financing ability effect is present, compared to only 42.85% (6 out of 14) for bank-driven countries. As I expected, these results indicate that the financing ability effect is more present in market-driven countries than in bank-based countries. What I did not expect is that in some bank driven countries the financing ability effect would not be present. In market driven countries the stock market is more active than in bank-driven countries, listed firms in market driven countries will therefore have easier access to non-bank funding compared to listed firms in bank driven countries. In bank-driven countries firms have to rely more on the banking-system than on the stock market, and have less access to non-bank funding. This could be the reason why the financing ability effect is more present in market driven countries and not in bank-driven countries. In table 7 this is shown with a coefficient of 0.017 for accounts receivable and -0.021 for accounts payable for listed firms within market-driven countries. This tells us that in market-driven countries listed firms provide significantly more trade credit and use significantly less trade credit than unlisted firms at a 1% significance level, confirming the financing ability effect. Table 7 also shows that in bank driven countries listed firms use significantly less trade credit than unlisted firms (-0.028),

4_{Brazil, United Kingdom, Korea, the Netherlands, Sweden, Singapore, Thailand, Taiwan and the United States.} 5_{Austria, Belgium, Brazil, Finland, United Kingdom, Japan, Poland, Sweden, Thailand and the United States} 6

(22)

22 confirming for one part the financing ability effect, though, listed firms also provide less trade credit (-0.006) than unlisted firms which does not confirm the financing ability effect in bank driven countries.

To check whether listed firms in market-driven countries do indeed provide significantly more trade credit and use significantly less trade credit than listed firms in bank-driven countries I made a regression that compares the receivables and payables of listed firms in market-driven countries with listed firms in bank-driven countries. The results are shown in table 8.

Table 8 about here

Table 8 shows that listed firms located in market-driven countries provide on average 0.06 (significant at a 1% level) more trade credit than listed firms located in bank-driven countries. Also is shown that listed firms located in market-driven countries use on average 0.044 (significant at a 1% level) less trade credit than listed firms located in bank-driven countries. These results confirm that due to easier access to non-bank financing in market-driven countries, listed firms are able to provide more trade credit and use less trade credit than listed firms in bank-driven countries where there is a less active stock market and firms are not able to rely that much on non-bank funding. With these result I can reject hypotheses HC0 and HD0, which stated that the payables ratio and the receivables ratio would be equal between listed firms located in market driven countries and listed firms located in bank driven countries. Next to that, I can accept the alternative hypotheses HC1 and HD1 which stated that the receivables ratio (payables ratio) of listed firms in market driven countries would be higher (lower) compared to listed firms in bank driven countries

6. CONCLUSION

(23)

23 unlisted firms have easier access to other sources of funding besides bank, the possibility exists that this advantage will be visible in their trade credit policy. As listed firms will also have more financing abilities to fund investment opportunities than unlisted firms, they do not have to use trade credit as a cheap way of funding but can use it to boost sales and enhance relationships

Multiple regression results showed that listed firms have significantly more receivables and significant less payables than unlisted firms, hereby rejecting HA0 and HB0, and accept the alternative hypotheses HA1 and HB1. Both coefficients are acting in the way the financing ability effect predicted. This paper thus confirms that the financing ability effect is present.

(24)

24 VII. REFERENCES

Almeida, H., Campello, M., Weisbach, M., 2004, The cash flow sensitivity of cash, The journal of

finance, Vol. LIX, No. 4

Banerjee, S., Dasgupta, S., Kim, Y., 2004, Buyer-supplier relationships and trade credit, Working

paper

Biais, B., Gollier, C., 1997, Trade Credit and Credit Rationing, The Review of Financial Studies, Vol. 10, No. 4, p. 903-937

Bougheas, S., Mateut, S., Mizen, P., 2009, Corporate trade credit and inventories: New evidence of a trade-off from accounts payable and receivable, Journal of Banking & Finance, Volume 33, Issue 2, p. 300–307.

Brennan, M., Maksimovic, V., Zechner, J., 1988, Vendor financing, The journal of finance, vol. XLIII, no. 5

Burkart, M., Ellingsen, T., 2004, In-kind Finance: A theory of trade credit, The American Economic

Review, vol. 94, No. 3, p 569-590

Cunningham, R., 2005, Trade Credit and Credit Rationing in Canadian Firms, Economic Analysis (EA)

Research Paper Series, Catalogue no. 11F0027MIE — No. 036

Deloof, M., 2003, Does Working Capital Management Affect Profitability of Belgian Firms?, Journal

of Business Finance & Accounting, 30(3) & (4)

Demirgüç-Kunt, A., Levine, R., 1999, Bank-based and market-based financial systems: cross country comparisons, The world bank Development research group, policy research working paper

Emery, G., 1987, An Optimal Financial Response to Variable Demand, Journal of Financial and

Quantitative Analysis, Vol. 22, No. 2 June, 209 – 225.

Ferris, J., 1981, A Transactions Theory of Trade Credit Use, Quarterly Journal of Economics, Vol. 96, No. 2, May, 243-270.

Jain, N., 2001, Monitoring Cost & Trade Credit, Quarterly Review of Economics and Finance, Spring 41(s), p. 89 – 110.

Levine, R., 2002, Bank-Based or Market-Based Financial Systems: Which is Better? Working paper McMillan, J., Woodruff, C., 1999, Interfirm Relationships and Informal Credit in Vietnam, The

Quarterly Journal of Economics, Vol. 114, No. 4, November, 1285-1320.

Mian, S., Smith, C., 1992, Accounts Receivable Management Policy: Theory and Evidence, The

Journal of Finance, Vol. 47, No. 1 p. 169-200

Mian, S., Smith, C., 1994, Providing trade credit and financing receivables, Journal of applied

(25)

25 Molina, C., Preve, L., 2009, Trade Receivables Policy of Distressed Firms and Its Effect on the Costs of Financial Distress, Financial management, Vol. 38, issue 3, p. 663 – 386

Nadiri, M., 1969, The Determinants of Trade Credit in the U.S. Total Manufacturing Sector,

Econometrica, Vol. 37, No. 3, p. 408-423

Ng, C., Smith, J., Smith R., 1999, Evidence on the Determinants of Credit Terms Used in Interfirm Trade, The Journal of Finance, Vol. 54, No. 3, p. 1109-1129

Opler, T., Pinkowitz, L., Stulz, R., Williamson, R., 1999, The determinants and implications of corporate cash holdings, Journal of Financial Economics, 52 3-46

Pagano, M., Panetta, F, Zingales, L., 1998, Why do companies go public, The Journal of Finance, Vol. LIII, No. 1

Petersen, M., & Rajan, R. (1997), Trade credit: Theories and evidence, Review of Financial Studies 10 (3), 661–691

Rajan, R., Zingales, L., 1995, What Do We Know about Capital Structure? Some Evidence from International, The Journal of Finance, Vol. 50, No. 5, p. 1421-1460

Schwartz, R., 1974, An Economic Model of Trade Credit, Journal of Financial and Quantitative

Analysis, IX 643--57.

Schwartz, R., Whitcomb, D., 1978, Implicit Transfers in the Extension of Trade Credit, in K. E. Boulding, and T. F. Wilson, eds., Redistribution Through the Financial System: The Grants

Economics of Money and Credit, p. 191-208.

Tewolde, S., Von Eije., H., (2007), Selfishness or Co-operation: Trade Credit in Manufacturing Firms in Eritrea, Research in Accounting in Emerging Economies, vol. 7, p. 369-387

(26)

26 Table 1. Number of listed and unlisted firms and the receivables and payables ratio by country for the 2005 – 2010 period.

Receivables ratio is measured by dividing accounts receivable by total assets. Payables ratio is defined by accounts payable divided by total assets. Whether a firm is listed or unlisted is defined according to the Orbis Database version of April 2012. The P-values refer to the significance level for differences in both ratios between listed and unlisted firms.

Country Number of firms Receivable ratio Payable ratio

Listed Unlisted Total Listed Unlisted p-value Listed Unlisted p-value

Austria 20 24 44 0.163 0.114 0.000 0.108 0.098 0.231 Belgium 16 112 128 0.209 0.162 0.001 0.139 0.142 0.813 Brazil 28 14 42 0.159 0.120 0.001 0.076 0.069 0.477 China 397 17 414 0.097 0.207 0.000 0.117 0.264 0.000 Germany 98 225 323 0.162 0.132 0.000 0.091 0.081 0.006 Spain 12 169 181 0.212 0.237 0.138 0.142 0.140 0.868 Finland 21 28 49 0.193 0.100 0.000 0.090 0.078 0.088 France 74 263 337 0.201 0.263 0.000 0.145 0.209 0.000 UK 75 257 332 0.144 0.124 0.000 0.097 0.093 0.443 Greece 28 15 43 0.179 0.216 0.035 0.083 0.109 0.012 Italy 54 324 378 0.203 0.236 0.000 0.172 0.189 0.016 Japan 816 334 1150 0.230 0.251 0.000 0.146 0.190 0.000 Korea (south) 184 106 290 0.186 0.225 0.000 0.115 0.166 0.000 The Netherlands 16 75 91 0.169 0.244 0.000 0.095 0.084 0.264 Norway 14 29 43 0.134 0.149 0.254 0.078 0.099 0.021 Poland 13 48 61 0.186 0.200 0.390 0.091 0.171 0.000 Russia 70 60 130 0.098 0.116 0.020 0.084 0.095 0.075 Sweden 33 50 83 0.190 0.116 0.000 0.105 0.105 0.952 Singapore 33 23 56 0.154 0.142 0.386 0.137 0.114 0.044 Thailand 27 39 66 0.129 0.120 0.452 0.101 0.129 0.005 Taiwan 278 14 292 0.152 0.087 0.000 0.113 0.048 0.000 Ukraine 23 17 40 0.145 0.128 0.311 0.111 0.129 0.262

(27)

27 Table 2, Descriptive statistics for listed and unlisted firms for the period 2005 - 2010

All variables for listed and unlisted firms are winsored for both sides at 1%. RETA indicates the ratio accounts receivable to total assets. PATA is accounts payable to total assets. SIZE is the natural logarithm of operating revenue (sales). NITA is net income divided by total assets. CATA is the cash ratio, calculated by dividing cash and cash equivalents by total assets. DEBT is measured by total long term debt divided by total assets. INVST is the future investment rate of the firm, calculated by the relative change of next year’s fixed assets. RISK is the riskiness of the cash flows, measured by taking the standard deviation of net income of the previous and current year.

Variable Firm-year

observations Mean

Standard

deviation Median Minimum Maximum

LISTED FIRMS RETA 17460 0.1711 0.100 0.1622 0.003 0.488 PATA 17460 0.1151 0.081 0.0962 0.003 0.401 SIZE 17453 13.2581 1.435 13.0082 1.946 18.928 NITA 17460 0.0361 0.070 0.0362 -0.259 0.243 CATA 17460 0.1421 0.120 0.1102 0.002 0.595 DEBT 17460 0.1161 0.121 0.0832 0.000 0.553 INVST 17460 0.0751 0.250 0.0002 -0.406 1.439 RISK 17460 0.0271 0.044 0.0112 0.000 0.278 UNLISTED FIRMS RETA 13698 0.1961 0.144 0.1762 0.000 0.630 PATA 13698 0.1461 0.122 0.1172 0.000 0.570 SIZE 13698 12.5171 1.132 12.4472 9.147 15.842 NITA 13698 0.0431 0.102 0.0282 -0.235 0.563 CATA 13698 0.0681 0.102 0.0242 0.000 0.519 DEBT 13698 0.1101 0.182 0.0142 0.000 0.893 INVST 13698 0.0471 0.250 0.0002 -0.480 1.527 RISK 13698 0.0381 0.089 0.0122 0.000 0.638 1

Indicate significant differences between listed and unlisted firms at a 1% significance level based on a t-test

2

(28)

28 Table 3. Univariate test for the entire sample for the period 2005 – 2010.

LISTED is a dummy variable which equals 1 if a firm is listed and 0 is the firm is unlisted. RETA indicates the ratio accounts receivable to total assets. PATA is accounts payable to total assets.

Receivables ratio Payables ratio

Coefficient Standard Error P-value Coefficient Standard Error P-value LISTED -0.025 0.003 0.000 LISTED -0.031 0.003 0.000 CONSTANT 0.196 0.003 0.000 CONSTANT 0.146 0.002 0.000

(29)

29 Table 4, Correlations matrix for listed and unlisted firms for the period 2005-2010

LISTED FIRMS

RETA PATA SIZE NITA CATA INVST DEBT RISK

RETA 1 0.585 0.013 0.020 -0.150 -0.044 -0.173 -0.11 PATA 1 0.111 -0.07 -0.111 -0.017 -0.154 -0.074 SIZE 1 0.102 -0.127 -0.036 0.222 -0.049 NITA 1 0.113 0.185 -0.154 -0.281 CATA 1 0.135 -0.292 0.076 INVST 1 -0.066 -0.029 DEBT 1 0.080 RISK 1 UNLISTED FIRMS

RETA PATA SIZE NITA CATA INVST DEBT RISK

(30)

30 Table 5. Development of variables over time for the period 2005 – 2010

LISTED FIRMS

RETA PATA SIZE NITA CATA DEBT INVST RISK

2005 0.179 0.118 13.139 0.044 0.136 0.113 0.062 0.000 2006 0.181 0.122 13.201 0.050 0.134 0.111 0.069 0.024 2007 0.179 0.121 13.252 0.050 0.135 0.110 0.129 0.023 2008 0.160 0.107 13.327 0.015 0.136 0.120 0.040 0.039 2009 0.161 0.108 13.198 0.014 0.155 0.122 0.148 0.041 2010 0.164 0.115 13.431 0.040 0.157 0.119 0.000 0.033 UNLISTED FIRMS

RETA PATA SIZE NITA CATA DEBT INVST RISK

(31)

31 Table 6. Regression results for the entire sample 2005 -2010

All variables for listed and unlisted firms are winsored for both sides at 1%. The regression includes dummy variables for country, industry and year. LISTED is a dummy variable which equals 1 if a firm is listed and 0 is the firm is unlisted. RETA indicates the ratio accounts receivable to total assets. PATA is accounts payable to total assets. SIZE is the natural logarithm of operating revenue (sales). NITA is net income divided by total assets. CATA is the cash ratio, calculated by dividing cash and cash equivalents by total assets. DEBT is measured by total long term debt divided by total assets. INVST is the investment rate of the firm, calculated by the relative change of fixed assets. RISK is the riskiness of the cash flows, measured by taking the standard deviation of net income of the previous and current year. DYEAR, DCOUNTRY and DINDUSTRY are dummy variables for respectively years, countries and industries. The OLS regression is corrected for autocorrelation and heteroskedasticity with the Newey-West estimator.

Coëfficiënt Std. Error P-Value Coëfficiënt Std. Error P-Value

LISTED 0.010 0.003 0.001 LISTED -0.036 0.003 0.000 PATA 0.580 0.016 0.000 RETA 0.407 0.012 0.000 SIZE -0.004 0.001 0.000 SIZE 0.014 0.001 0.000 NITA 0.105 0.012 0.000 NITA -0.120 0.009 0.000 CATA -0.004 0.001 0.000 CATA -0.007 0.009 0.386 DEBT -0.059 0.007 0.000 DEBT -0.030 0.006 0.000 INVST 0.015 0.003 0.000 INVST 0.004 0.002 0.066 RISK -0.107 0.015 0.000 RISK 0.079 0.011 0.000 CONSTANT 0.146 0.022 0.000 CONSTANT -0.040 0.026 0.124

DYEAR YES DYEAR YES

DCOUNTRY YES DCOUNTRY YES

DINDUSTRY YES DINDUSTRY YES

(32)

32 Table 7. A comparison between market and bank driven countries

The coefficients for the receivables and the payables ratio indicate the difference between listed and unlisted firms for each country. A positive coefficient means that listed firms have a higher ratio than unlisted firms. The p-value indicates the significance of the difference between listed and unlisted companies.

Country

Firm-year

observations Coefficient Std. Error

P-value Coefficient Std. Error

P-value Austria 264 0.064 0.021 0.003 -0.043 0.017 0.014 Belgium 768 0.046 0.016 0.004 -0.051 0.017 0.003 Brazil * 252 0.059 0.019 0.002 -0.004 0.011 0.694 China 2484 -0.056 0.028 0.049 -0.114 0.029 0.000 Germany 1938 0.011 0.009 0.220 -0.001 0.006 0.945 Spain 1086 -0.046 0.023 0.046 -0.021 0.016 0.204 Finland 294 0.067 0.011 0.000 -0.031 0.007 0.000 France 2020 -0.011 0.008 0.150 -0.051 0.006 0.000 UK * 1992 0.020 0.005 0.000 -0.016 0.004 0.000 Greece 257 -0.037 0.018 0.045 -0.023 0.010 0.019 Italy 2268 -0.009 0.009 0.295 -0.031 0.007 0.000 Japan 6900 0.021 0.002 0.000 -0.038 0.002 0.000 Korea (south) * 1740 0.001 0.004 0.817 -0.038 0.004 0.000 The Netherlands * 546 -0.077 0.018 0.000 0.002 0.011 0.861 Norway 257 -0.006 0.016 0.692 -0.022 0.011 0.043 Poland 366 0.024 0.012 0.051 -0.090 0.013 0.000 Russia 780 -0.003 0.008 0.706 -0.020 0.007 0.002 Sweden * 498 0.080 0.009 0.000 -0.008 0.008 0.281 Singapore * 336 0.014 0.015 0.325 0.018 0.012 0.126 Thailand * 396 0.022 0.011 0.039 -0.036 0.008 0.000 Taiwan * 1752 0.005 0.008 0.568 0.001 0.008 0.854 Ukraine 240 -0.006 0.014 0.649 -0.009 0.014 0.520

The United States * 3615 0.069 0.006 0.000 -0.013 0.004 0.002

Market driven countries 11127 0.017 0.002 0.000 -0.021 0.002 0.000

Bank driven countries 20024 -0.006 0.002 0.001 -0.028 0.001 0.000

(33)

33 Table 8. A comparison between market and bank driven countries for listed firms.

All variables for listed and unlisted firms are winsored for both sides at 1%. MARKET is a dummy variable which equals 1 if a listed firm is located in a market-driven country and 0 is the listed firm is located in a bank-driven country RETA indicates the ratio accounts receivable to total assets. PATA is accounts payable to total assets. SIZE is the natural logarithm of operating revenue (sales). NITA is net income divided by total assets. CATA is the cash ratio, calculated by dividing cash and cash equivalents by total assets. DEBT is measured by total long term debt divided by total assets. INVST is the investment rate of the firm, calculated by the relative change of fixed assets. RISK is the riskiness of the cash flows, measured by taking the standard deviation of net income of the previous and current year. DYEAR, DCOUNTRY and DINDUSTRY are dummy variables for respectively years, countries and industries. The OLS regression is corrected for autocorrelation and

heteroskedasticity with the Newey-West estimator.

Coefficient Std. Error

P-value Coefficient Std. Error P-value MARKET 0.060 0.014 0.000 MARKET -0.044 0.011 0.000 PATA 0.649 0.017 0.000 RETA 0.493 0.015 0.000 SIZE -0.009 0.001 0.000 SIZE 0.011 0.001 0.000 NITA 0.122 0.013 0.000 NITA -0.122 0.011 0.000 CATA -0.137 0.009 0.000 CATA -0.013 0.008 0.102 DEBT -0.085 0.009 0.000 DEBT -0.027 0.007 0.000 INVST 0.009 0.003 0.001 INVST 0.003 0.002 0.152 RISK -0.058 0.021 0.005 RISK 0.024 0.018 0.193 CONSTANT 0.185 0.018 0.000 CONSTANT -0.094 0.015 0.000

DYEAR YES DYEAR YES

DCOUNTRY YES DCOUNTRY YES

DINDUSTRY YES DINDUSTRY YES

Observations: 17453 Observations: 17453