TRADE CREDIT AND SME PROFITABILITY. Preliminary draft

Transcription

1 150b TRADE CREDIT AND SME PROFITABILITY Preliminary draft Cristina Martínez-Sola Dep. Accounting and Finance Faculty Social Sciences and Law University of Jaén Jaén (SPAIN) Pedro J. García-Teruel Dep. Management and Finance Faculty of Economics and Business University of Murcia Murcia (SPAIN) Pedro Martínez-Solano Dep. Management and Finance Faculty of Economics and Business University of Murcia Murcia (SPAIN) Área temática: b) valoración y finanzas Keywords: accounts receivable, trade credit, profitability, SMEs. JEL Classification: G30, G31 1

2 TRADE CREDIT AND SME PROFITABILITY Abstract Financial literature has discussed in depth motives for trade credit provision by suppliers. However there is no empirical evidence on the effect of granting trade credit on firm profitability. Trade credit has an effect on the level of investment in current assets and consequently may have an important impact on the profitability and liquidity of the firm. We examine the profitability implications of providing financing to customers for a sample of 11,337 Spanish manufacturing SMEs during the period This paper also explains the differences in the profitability of trade credit according to financial, operational, and commercial motives. The findings suggest that managers can improve firm profitability by increasing their investment in receivables, and that effect is greater for larger, more liquid firms, firms with volatile demand, and for firms with more market share. 2

3 1. INTRODUCTION Trade credit is an arrangement between a buyer and seller by which the seller allows delayed payment for its products (Mian and Smith, 1992), instead of cash payment. According to Lee and Stowe (1993), it is part of a joint commodity and financial transaction in which a firm sells goods or services and simultaneously extends credit for the purchase to the customer. Trade credit plays an important role in firm financing policy. For the buyer, it is a source of financing through accounts payable, while for the seller, trade credit is an investment in accounts receivable. We focus on the supply side of trade credit. There have been several sorts of explanations proposed for trade credit. The financial motive (Emery, 1984; Mian and Smith, 1992; Schwartz, 1974) argues that firms able to obtain funds at a low cost will offer trade credit to firms facing higher financing costs. Emery (1984) sees trade credit as a more profitable short-term investment than marketable securities. The operational motive (e.g. Emery, 1987) stresses the role of trade credit in smoothing demand and reducing cash uncertainty in the payments (Ferris, 1981). And, according to the commercial motive, trade credit improves product marketability (Nadiri, 1969) by making it easier for firms to sell. Trade credit can also be used to maximize profit through price discrimination (Brennan, Maksimovic, and Zechner, 1998). Finally, according to the product quality motive (e.g. Smith, 1987), firms extend trade credit to guarantee product quality, by alleviating information asymmetry between buyers and sellers. Previous studies have focused on explaining the determinants of trade credit (Cheng and Pike, 2003; Deloof and Jegers, 1996, 1999; Elliehausen and Wolken, 1993; Garcia- Teruel and Martinez-Solano, 2010; Hernandez and Hernando 1999; Long, Malitz, and Ravid, 1993; Ng, Smith, and Smith, 1999; Niskanen and Niskanen, 2006; Petersen and Rajan, 1997; Pike, Cheng, Cravens, and Lamminmaki, 2005; Wilson and Summers, 2002; among others), with most of this literature focused on large firms. However, trade credit is particularly important in the case of small and medium-sized companies, since trade debtors are the main asset on most of their firms balance sheets. Giannetti (2003) provides details on firm balance sheets in eight European countries. It is noteworthy that Spain presents the second highest ratio of trade debtors to total assets (35%), after Italy (42%), and holds more than a third of its invested assets in trade credit, while the countries with less reliance on trade credit are UK (20.47%), and Netherlands (13.28%). 3

4 Given the significant investment in accounts receivable, the choice of credit management policies could have important implications for firm profitability. Though the impact of trade credit policy on profitability and value could be highly important, no studies have been carried out to examine this relationship. The only exception is Hill, Kelly, Lockhart, and Washam (2010), who study the shareholder wealth implications of corporate trade credit policy but for a sample of large US firms. Therefore, the purpose of this study is to find empirical evidence of the effect of granting trade credit on firm profitability. In order to do this we set up a panel of 11,337 Spanish small to medium-sized enterprises during the period 2000 to This research contributes to the financial literature in several ways. First, we test the relation trade credit-profitability for a sample of Spanish SMEs because of their particular institutional setting, which makes Spain a country where trade credit is particularly important. Proof of this is that Spanish firms have one of the longest effective credit periods in Europe (Marotta, 2001), thereby providing an excellent context in which to study the implications of trade credit profitability. Secondly, to the greater importance of trade credit for SMEs, because of problems of asymmetric information and their greater difficulty in accessing capital markets (Petersen and Rajan, 1997; Berger and Udell, 1998), should be added the fact that the Mediterranean countries (such as Spain) have a greater preponderance of smaller firms than northern European or Scandinavian countries (Mulhern, 1995). So, we also contribute to the SME literature. Regarding the institutional setting, Spain is a civil-law country, characterized by its less developed capital markets (La Porta, Lopez-de-Silanes, Shleifer, and Vishny, 1998), lack of creditor protections, and weak legal system. Studies such as Demigurc-Kunt and Maksimovic (2002) argue that firms operating in countries with more developed banking systems grant more trade credit to their customers. In countries with weak legal systems the provision of trade credit by suppliers may also be an important channel by which firms can access capital indirectly, through their suppliers (Frank and Maksimovic, 2001), because of the difficulty in accessing financial markets (La Porta et al., 1998; Demirguc- Kunt and Maksimovic, 1998; and Rajan and Zingales, 1998). An additional factor that can make trade credit use in Spain more intensive than in other countries is the weakness of the judicial system (San-Jose and Cowton, 2009), since there is no government implication in reducing the period to pay suppliers. This study obtains empirical evidence of a lineal relationship between trade receivables and profitability of SMEs, which implies that the benefits of supplier financing overcome 4

5 the costs associated with trade credit. This relationship is maintained when firms are classified according to their activity and with different estimation methods. Further evidence supports the financial motive for trade credit; larger and more liquid firms (financially unconstrained firms) obtain extra profitability by granting trade credit than do smaller and less liquid firms (financially constrained firms). The findings also support the operational motive for trade credit; for firms with uncertain demand, the use of trade credit is more profitable than for firms with certain demand. Trade credit might be used to smooth demand, thus enhancing firm profitability. We do not find evidence that trade credit as an instrument in mitigating information asymmetry regarding product quality increases a firm s profitability. On the contrary, larger companies get more return on investment in trade credit than smaller firms with no reputation in product markets. Finally, firms with more market share obtain higher profitability from trade credit. Our results indicate that certain firms are better able to derive strategic benefits from credit policy. The remainder of the article is organized as follows. In the next section, we review previous trade credit literature and discuss predictions on the relations between the supply of trade credit and firm profitability. Section 3 describes the sample and variables. In section 4 we present the summary statistics of the variables employed in the study. Section 5 specifies the model to estimate. In section 6 we report the results and section 7 concludes. 2. REVIEW OF LITERATURE Trade credit-profitability relationship The first research question we try to answer is whether trade credit increases profitability. There are many reasons that lead suppliers to extend credit. Chiefly, granting trade credit enhances firm s sales, and consequently may result in higher profitability. Meltzer (1960) states that a primary function of trade credit is to mitigate customers financial frictions, thus facilitating increased sales and market share growth (Nadiri, 1969). In addition to resolving financing frictions, trade credit can boost sales by alleviating informational asymmetry between suppliers and buyers in terms of product quality (Long et al., 1993, Smith, 1987). In this sense, the seller s investment in trade credit facilitates exchange by reducing uncertainty about product quality. Also, trade credit enables price discrimination (Brennan et al., 1998); by varying the period of credit 5

6 or the discount for prompt payment, firms can sell their products at different prices depending on the demand elasticity of customers. In a long term perspective, trade credit might give future profits by establishing and maintaining permanent commercial relationships (Ng et al., 1999; Wilner, 2000). Besides increased sales, trade credit may increase revenues through interest income (Emery, 1984) or reduction in transaction costs (Ferris, 1981; Emery, 1987). However, the provision of trade credit entails negative effects such as default risk or late payment, which may damage firm profitability. Moreover, extending supplier financing involves administrative costs associated with the granting and monitoring process, as well as transaction costs for converting receivables into cash (Emery 1984). Further, carrying receivables on the balance sheet implies direct financing and opportunity costs, so reducing funds available for expansion projects. Theoretical models argue that there is an optimal trade credit policy (Nadiri, 1969; Emery, 1984), where the optimal level of accounts receivable occurs when the marginal revenue of trade credit is equal to the marginal cost (Emery, 1984). On the other hand, Lewellen, McConnell and Scott (1980) demonstrated that trade credit can be used to increase firm value when financial markets are imperfect. Consequently, one might expect a quadratic relationship between trade credit and firm value or profitability determined by a tradeoff between costs and benefits of supplying trade credit, where there is a level of trade credit granted which maximizes firm value or profitability. Moreover, these theoretical models do not find empirical support. Actually, Hill et al. (2010) find a lineal relationship between trade credit and firm value, where the benefits of granting trade credit surpass the costs. This effect may be even greater in the case of SMEs. Cheng and Pike (2003) find that firms operating in competitive markets are forced to offer industry credit terms. In effect, SMEs are forced to grant trade credit despite the costs associated to it, because not granting trade credit would lose sales, and profitability would decrease. We therefore expect a lineal relationship between the investment in trade credit and profitability. Trade credit, firm characteristics and profitability In this section, we analyze the effect of firm characteristics on trade credit profitability. We review the motives and incentives for trade credit extension in the financial literature, and we establish the expected impact of trade credit on profitability. The supplier firm's motives for offering trade credit can be classified as financial, operational and commercial. 6

7 Schwartz (1974) developed the financial motive for the use of trade credit. He suggests that when credit is tight, financially stable firms will increasingly offer more trade credit to maintain their relations with smaller customers, who are rationed from direct credit market participation. The seller firm acts as a financial intermediary to customers with limited access to capital markets, financing their customers growth. Petersen and Rajan (1997) find empirical evidence that firms with better access to capital markets offer more trade credit. Larger firms are thought to be better known and have better access to capital markets than smaller firms, in terms of availability and cost, and should therefore face fewer constraints when raising capital to finance their investments (Faulkender and Wang, 2006). Financial motive predicts a positive connection between extending trade credit and firm size according to which, creditworthy firms should extend trade credit to less creditworthy firms (Emery, 1984; Mian and Smith, 1992; Schwartz, 1974). According to the financial motive of trade credit, we expect a greater effect of trade credit on firm profitability for the subsample of larger firms. Emery's paper (1984) is based on information costs. Capital market imperfections require selling firms to maintain adequate liquid reserves that they either can invest in marketable securities or lend out through trade credit. Imperfections also allow seller firms to acquire knowledge about customers' ability to pay at a relatively low cost. This creates an informational advantage over third party intermediaries and allows sellers to offer trade credit at an implicit interest rate that is lower than the purchaser could obtain elsewhere. In this sense, Emery (1984) argues that suppliers may extend credit if the implicit rate of return 1 earned on receivables exceeds that of other investments. Petersen and Rajan (1994) and Atanasova (2007) show that implicit returns earned from trade credit are typically large, relative to feasible opportunity costs. The Emery model (1984) suggests that more liquid firms will extend trade credit as an alternative to investing in marketable securities. In the same vein, Ng et al. (1999) argue that trade credit is given from firms with high liquidity to firms with low liquidity. Consequently, we expect that more liquid firms secure a higher return on investment in trade credit. The financial motive for trade credit implies that larger, more financially secure producers will offer trade credit to their smaller customers. Large firms extend trade credit to their customers in order to secure repeat sales and to build long-term relationships. However, from the standpoint of commercial motive, smaller firms that have worse reputations 1 In trade credit arrangements it is very common to offer early payment discount to the customer. The most common payment term is 2/10, net 30 (Ng et al., 1999), by which a customer takes 2 percent discount on the purchase price if the payment is made within ten days; otherwise the payment is in full within thirty days. This translates as over 40 percent annual rate. 7

8 need to use more trade credit in order to guarantee their products (Long et al., 1993), which contradicts the predictions of financial motive for trade credit. From this perspective, a higher effect of trade credit on firm profitability for smaller firms might be expected. Emery (1987) focuses on trade credit as an operational tool, addressing the role of uncertain product demand in a firm's operating decisions. As demand fluctuates, sellers face two alternatives: either they can allow the selling price to fluctuate so that the market always clears, or they can vary production to match demand. Either option is quite costly. If price varies, potential buyers face extremely high costs of information search. If production varies, sellers face extremely high production costs. Trade credit could help to smooth irregular demand through stimulating sales by relaxing trade credit terms in slack demand periods (Emery 1984, 1988; Nadiri, 1969). The operational motive predicts that firms with variable demand extend significantly more trade credit than firms with stable demand. Long et al. (1993) find empirical evidence that is consistent with this view. We test the effect of trade credit under uncertain product demand conditions on firm profitability. Following the operational motive, we expect the profitability of receivables held by firms with high sales uncertainty to be higher than that held by firms with sales certainty. Lastly, from a commercial perspective, Nadiri (1969) argues that availability of alternative payment terms can expand the market by increasing product demand. According to the commercial motive, trade credit improves product marketability by facilitating firm s sales. So, for firms with less market share (less market power) trade credit should prove more beneficial, as these firms have stronger incentives to increase sales (Hill et al., 2010). Hill et al. (2010) find that the profitability of receivables is a decreasing function of market share. However, market pressures might force small business with no market power to offer normal industry credit terms, regardless of its possible negative impact on profitability. We test the effect of trade credit on profitability for less market presence firms and for firms with high market share. 8

9 3. DATA AND VARIABLES Data The data used in this study were obtained from the SABI database. This database contains financial and economic data on small and medium-sized Spanish firms. According to the requirements established by the European Commission s recommendation 2003/361/CE of 6th May 2003, small and medium-sized firms are those meeting the following criteria for at least three years: fewer than 250 employees; turnover of less than 50 million; and less than 43 million in total assets. In addition, a series of filters was applied. Thus, the observations of firms with anomalies in their accounts were eliminated, for example negative values in their assets or sales, and firms whose total assets differ from total liabilities and equity. Finally, to reduce the impact of outliers, the variables employed in this paper are winsorized at 1% and 99% level. The final sample consists of an unbalanced panel of 71,635 firm-year observations for 11,337 manufacturing companies over the period. We chose a sample of manufacturing firms due to the homogeneity across industries in credit terms. Variables Return on assets (ROA) is used as the dependent variable to analyze the effect of trade credit on a firm s profitability. This variable is defined as the ratio of operating income to total assets, or EBIT to total assets (Michaelas, Chittenden, and Panikkos, 1999; Titman and Wessels, 1988). The key independent variable is the investment in accounts receivable; REC, it is the ratio of accounts receivable to total assets (Deloof and Jegers, 1999; Cuñat, 2007; Boissay and Gropp, 2007). We also employ an additional variable to take into account industry differences - ADJUSTEDREC, which is firm accounts receivable minus industry mean accounts receivable. Because of the benefits of trade credit, we expect a positive relationship with ROA, for both measures of receivables. In order to analyze the effect of varying industry terms, we define ARDEVIATION, a dummy variable that takes value one for firms granting shorter trade credit periods than the industry mean. We expect a negative sign for this variable, indicating that shorter credit periods than the industry mean reduce firm profitability. 9

10 All regressions include control variables found by previous authors to explain either trade credit or firm profitability (i.e. Deloof, 2003): size of the firm, growth in its sales, and its leverage. SIZE is the logarithm of total assets. There is no consensus about the relation between value and size of the firm. For instance, Lang and Stulz (1994) find a negative relation between firm size and performance for U.S. companies, whereas Berger and Ofek (1995) find a positive relation. So, we cannot establish a clear relationship between firm size and profitability. GROWTH is sales annual growth (Sales t Sales t-1 )/Sales t-1. In this sense, Scherr and Hulburt (2001) assume that firms that have grown well so far are better prepared to continue to grow in the future. Thus we expect a positive relationship between growth opportunities and firm profitability. Finally, DEBT is the ratio of debt to total assets. Theory points in different directions with respect to the impact of debt on firm profitability (Harris and Ravid, 1991). Debt may yield a disciplinary effect when free cash flow exists (Jensen, 1986; Stulz, 1990). Also, firms can use debt to create tax shields (Modigliani and Miller, 1963). However, debt can increase conflicts of interest about risk and return between creditors and equity holders (Joh, 2003). Thus, firms may use less debt to avoid external finance costs (Myers and Majluf, 1984). The greater information asymmetry and agency conflicts associated with debt for smaller firms could lead creditors to demand higher return (Pettit and Singer, 1985). Therefore, we expect a negative effect of debt on profitability. Furthermore, and since good economic conditions tend to be reflected in a firm s profitability, controls were applied for the evolution of the economic cycle using the variable GDP, which measures annual GDP growth. To test the financial motive for trade credit we split the sample according to firm size, measured as DSIZE - a dummy variable that takes the value one if SIZE is less than or equal to the median firm size in the sample, and zero otherwise - and liquidity measured as DLIQ - a dummy variable that takes the value one if firm liquid assets are smaller than or equal to the median liquid assets. For unlisted companies the size of the firm is a common proxy for financial constraints (Almeida, Campello, and Weisbach, 2004; Faulkender and Wang, 2006) or creditworthiness (Petersen and Rajan, 1997). From another point of view, firm size is often a proxy for reputation for product quality (Long et al., 1993). In this sense, smaller firms are less likely to have established reputations (Berger and Udell, 1998). As stated in the previous section, we expect larger firms to have greater profitability from receivables than smaller firms. To test the effect of the operational motive for trade credit on firm profitability we split the sample according to SALESVOL a variable reflecting demand variability. Following 10

11 Long et al. (1993), it is the standard deviation of sales (three years) divided by mean sales over a three-year period. DSALESVOL is a dummy variable that takes the value one if SALESVOL is smaller than or equal to the median sales volatility in the sample. We expect a greater effect of trade credit on firm profitability for the subsample of uncertain product demand. Finally, to test the commercial motive for trade credit we split the sample according to firm market share. We define DMKSHARE as a dummy variable that takes the value one if MKSHARE is smaller than or equal to the median market share in the sample, where MKSHARE is the ratio of annual firm sales to annual industry sales. Due to the possible existence of two opposing effects, we cannot establish a clear relationship between market share and profitability of receivables. 4. SUMMARY STATISTICS Table 1 offers descriptive statistics of the variables employed in this paper. The return on assets is around 6.5 percent. The economic importance of trade credit is obvious. Consistent with the study of Giannetti (2003), we find that, for the average company, accounts receivable represents the largest asset category on the balance sheets; the investment in accounts receivable is over 34 percent of total assets and the number of days for accounts receivable is around 97 days. Together with this, the average firm has growth sales of 9 percent, and 64 percent of their total liabilities and shareholders equity is taken up by debt. In the period analyzed ( ), the GDP grew at an average rate of 4.05 percent in Spain (expansionary phase of business cycle). Table 1 here Table 2 shows the correlation matrix for the variables defined above. There is a significant positive correlation between the return on assets and accounts receivable to assets (0.1595) and between the return on assets and accounts receivable adjusted by industry (0.1611). This shows that the supply of trade credit is associated with an increase in the firm s profitability. As regards control variables, SIZE is positively related to ROA, although the correlation is very small (0.0099). There is a significant positive correlation between GROWTH and ROA (0.2071), while DEBT is negatively correlated with ROA ( ). With regard to the correlations between independent variables, 11

12 there are no high values between them, which could lead to multicolineality problems and, consequently, inconsistent estimations. Table 2 here Table 3 reports return on assets and accounts receivable by sector of activity. Ng et al. (1999) find that trade credit practice is likely to show wide variation across industries in credit terms, but little variation within industries. Thus, we split the sample according to NACE (Rev. 2) two digits code (10-33), resulting in a total of 24 industries. Manufacture of beverages is the industry with the lowest investment in receivables with a value of 24 percent, followed by manufacturing of food products, and manufacturing of furniture with an investment in receivables of 29 percent. This result is not surprising, as these industries heavily rely on cash sales. In contrast, firms dedicated to the manufacture of electrical equipment, manufacture of computer, electronic and optical products, and repair and installation of machinery and equipment have the highest ratio of receivables over assets, with an average ratio of 39.5 percent. We find that differences in the means are statistically significant (ANOVA test). Figure 1 shows the REC-ROA relation by industry. Graphically, we can see a positive linear relationship between accounts receivable and the return on assets. Overall, higher levels of trade credit are related to better profitability. Table 3 here Figure 1 5. MODEL SPECIFICATION In order to check for a linear relation between trade credit (REC) and profitability (ROA), we estimate equation (1). Next, to check for a non linear relation between REC and profitability ROA, we also incorporate REC 2 into the model. ROA it = ß 0 + ß 1 REC it + ß 2 SIZE it + ß 3 DEBT it + ß 4 GROWTH it + ß 5 GDP it +? i +? t +? it (1) where ROA is firm profitability, REC is receivables to assets ratio, SIZE is firm size, DEBT is leverage, GROWTH is growth opportunities, and GDP is annual GDP growth. 12

13 With the aim of examining the different effect of trade credit on profitability, from financial, operational and commercial motives, we analyze the impact of firm characteristics on the value of receivables by including dummy variables. Thus, the model to estimate is as follows: ROA = ß 0 + (ß 1 + ß 2 DUMMY) REC it + ß 3 DUMMY + ß 4 SIZE it + ß 5 DEBT it + ß 6 GROWTH it + ß 7 GDP it +? i +? t +? it (2.1) Or: ROA = ß 0 + ß 1 REC it + ß 2 REC it DUMMY + ß 3 DUMMY + ß 4 SIZE it + ß 5 DEBT it + ß 6 GROWTH it + ß 7 GDP it +? i +? t +? it (2.2) where DUMMY take values 0 and 1 and, specifically, will be DSIZE, DLIQ, DSALESVOL, and DMKSHARE when we study the financial market access, liquidity, sales volatility, and market share, respectively. If we differentiate firm profitability (ROA) with respect to the investment in trade credit (REC), we obtain ß 1 + ß 2 DUMMY. Consequently, if DUMMY takes value 1, then the effect of REC on ROA is explained by ß 1 + ß 2. If DUMMY takes value 0, then the effect is explained by ß TRADE CREDIT AND FIRM PROFITABILITY Our base method of estimating is Ordinary Least Squares (OLS). Next, we introduce a fixed effect estimation (FE) to control for the presence of individual heterogeneity. Fixed effects estimation assumes firm specific intercepts, which capture the effects of those variables that are particular to each firm and that are constant over time. Finally, to control for the potential endogeneity problem that may exist if trade credit policy correlates with unobservable heterogeneity influencing firm s profitability, we use instrumental variables estimation 2. Trade Credit-Profitability Relationship In order to analyze the evolution of firm profitability in function of the investment in receivables we perform a univariate analysis. In Table 4, we present the mean values of 2 We use as instruments the variable REC lagged 1 period. 13

14 ROA variable for each decile of the variable REC. We observe greater profitability by firms with more trade credit investment. Figure 2 shows levels of ROA according to REC. This suggests a linear relation between trade credit and profitability, as we can see higher investment in trade credit is related to better profitability. However, the results obtained in this analysis are not sufficient to describe the relation between trade credit and firm profitability, so we conduct further analysis. Table 4 here Figure 2 In table 5, columns 1, 3 and 5, we present the results of the estimation of our initial model (Equation 1) for the full sample. The stand-alone coefficient on REC is positive and significant. As previously defined, REC is the investment in accounts receivable, calculated as the ratio of accounts receivable to total assets. In columns 2, 4 and 6 we introduce a quadratic term to test the existence of a nonlinear relation between receivables and ROA. REC 2 is not statistically different from zero, while the coefficient is positive for the instrumental variables method of estimation. This implies a positive relation between trade credit and firm profitability, the supply of trade credit is beneficial despite the existence of credit management costs, as well as late payment and exposure to payment default. This relation is maintained when firms are classified according to their activity and for three methods of estimation. The quadratic relationship is not found in our sample. This behavior is inconsistent with the tradeoff between costs and benefits of receivables, although other explanations are possible. Cheng and Pike (2003) find that firms operating in competitive markets are forced to offer industry credit terms. This argument could be extended since SMEs could be forced to grant trade credit in spite of the costs associated, because not granting trade credit would lose sales and profitability would decrease. In addition, Hill et al. (2010) find empirical evidence of a positive lineal relationship between receivables and firm value. Table 5 here In table 6, we insert in equation (1) an alternative measure for trade credit adjusted by industry, which replaces REC, in order to take into account industry differences. This is ADJUSTEDREC, measured as firm account receivables minus industry mean account 14

15 receivables. Since ADJUSTEDREC variable is also positively and significantly related to ROA (columns 1, 3 and 5), we obtain additional empirical evidence of a linear relation between trade credit and firm profitability. Several authors, such as Smith (1987), Ng et al. (1999) and Fisman and Love (2003) find that trade credit terms are uniform within industries and differ across industries. Smith (1987) argues that within an industry both parts, buyers and sellers, face similar market conditions, while across industries market conditions and investment requirements in buyers may vary significantly. Paul and Boden (2008) suggest that firms need to match normal industry terms to maintain their market competitiveness. If the credit granted by a firm is not competitive compared to firms in the same sector, this could have negative effects on firm profitability. In order to analyze the effect of varying industry terms we define ARDEVIATION, a dummy variable that takes value one for firms granting shorter trade credit periods than the industry mean. We test the hypothesis that differences in trade credit period related to industry terms lead to worse operating profitability. Columns 2, 4 and 6 incorporate ARDEVIATION and REC ARDEVIATION in equation 1. REC ARDEVIATION coefficient is negative, but for fixed effects estimation, which indicates that shorter credit periods than the industry mean reduced firm profitability. This is consistent with Hill et al. (2010), who assert that may be longer credit periods help customers facing liquidity problems, which may facilitate future sales. Table 6 here Trade credit, firm characteristics and profitability To test the hypothesis put forward in our paper that the relation between firm profitability and trade credit differs according to firms characteristics, we develop a model that relates firm profitability to trade credit, incorporating interaction terms between the receivables ratio and dummy variables measuring size, liquidity, sales volatility, and market share (equation 2). In the odd columns (1, 3, and 5) of table 7, we test the effect of firm size on the profitability of receivables. To empirically estimate the effect on profitability of the supply of trade credit depending on firm s size, we regress firm s profitability (ROA) on receivables (REC), a dummy variable that takes the value one if the firm size is smaller than the median value and zero otherwise (DSIZE), and their cross-effect (REC DSIZE). 15

16 Simillary to Dittmar and Mahrt-Smith (2007) we also include in the model variable DSIZE by itself. If an endogenous relation exists, it is more likely to show up in the dummy variable than in the interaction with receivables. The REC DSIZE negative interaction coefficient indicates that trade credit investment is more profitable for larger firms than for smaller firms, except for fixed effect estimation for which there are no significant differences in the profitability of receivables according to firm size. For instance, with the OLS method of estimation, the profitability of receivables for the subsample of smaller firms (DSIZE=1) is (-0.011) = , while for the subsample of larger firms this value is Similarly for instrumental variables estimation; for larger firms and for smaller firms. This result is consistent with the view that unconstrained firms (larger firms) offer trade credit to finance their customer s growth because of their greater financial capacity, so increasing profitability. The limited level of financial development together with the small size of the firms in the sample makes it more difficult and expensive for firms to access capital markets. Trade credit implies accounts receivable financing, since it requires the seller to seek financing from a third party, such as a bank. Thus, smaller firms obtain less profitability from receivables than larger firms. These results support the financial motive for trade credit and are not consistent with product quality guarantee argument. Moreover, other reasons explaining this higher return could be the existence of scale economies associated to trade credit management in larger firms (administrative costs associated with the granting and monitoring process), more efficient credit management (Peel, Wilson, and Howorth, 2000), and better capacity to enforce contracts. In the even columns (2, 4 and 6) of table 7, we consider that the liquidity of the firm may affect the value of accounts receivable. In this sense, Petersen and Rajan (1997) find that suppliers offering product financing have excess liquidity, which is consistent with trade credit serving to mitigate buyers financial constraints. We examine the liquidity effect on the value of accounts receivable by including the REC DLIQ variable. Since the interaction variable coefficient? 1 is negative and significant, the sum of the coefficients ß 1 + ß 2 is lower than ß 1, indicating that the profitability of receivables is lower for the subsample of less liquid firms (DUMMY=1). We find that liquidity is a factor which positively affects the profitability of receivables. Trade credit helps to mitigate customers financial frictions. Moreover this positive effect can be explained by the fact that the implicit return on receivables is greater than the return on alternative investment, as Petersen and Rajan (1994) and Atanasova (2007) find. 16

17 Table 7 here. According to the operational motive, trade credit incentivizes customers to acquire merchandise at times of low demand (Emery, 1987). Long et al. (1993) find a direct relation between trade credit levels and demand uncertainty. We try to test the effect of operational motive on the profitability of receivables including an interaction variable REC DSALESVOL. In columns 1, 3 and 5 of table 8, we report the results. For the subsample of firms with more sales volatility or uncertain demand (DUMMY=0), the effect of REC on ROA is determined by the coefficient ß 1 (variable REC), while for the subsample of firms with less sales volatility or certain demand (DUMMY=1), the effect is explained by ß 1 + ß 2 (variables REC + REC DLIQ). Since the interaction coefficient? 1 is negative and statistically significant, the profitability of receivables for firms with uncertain demand is higher than for firms with a stable demand. The slope of the function is higher for firms with uncertain demand than for certain demand. This finding supports the operational motive for trade credit. The negative effect of the variable REC DSALESVOL on firm profitability may be a result of costs reduction for firms with uncertain demand. Trade credit policy can be used to mitigate the consequences of uncertain sales (Emery, 1987). In columns 2, 4, and 6 of table 8, we estimate model 2 including DMKSHARE and REC DMKSHARE to get additional information about the effect of trade credit to stimulate sales, and consequently enhance profitability. The results in columns 1 and 5 indicate that for firms with greater market presence the supply of trade credit than for firms with smaller market shares is more profitable, since REC DMKSHARE is statistically significant and negatively signed. In column 3, we do not find differences in the profitability of receivables between less market share and more market share firms. Unlike Hill et al. (2010), we do not find evidence that the incentives to extend financing are reduced for firms with larger market shares. The reasons are various. On the one hand, firms with market power are not forced to grant trade credit in the same way as firms with less market presence, so these firms will evaluate credit risks and grant trade credit to their customers with higher credit quality. On the other, firms with market power are better able to enforce contracts and may suffer less debt defaults. This result further supports financial motive, as larger firms are likely to have greater market shares. Table 8 here. 17

18 Regarding control variables results, GROWTH is positive in all cases, so any increase in sales causes profit to grow. Moreover, sales growth could be an indicator of a firm s investment opportunities, and it is an important factor in allowing firms to enjoy improved profitability. Consistent with Myers` pecking order theory, we find that profitability is negatively related to debt. However, we find contradictory empirical evidence of the relationship between size and profitability. Overall, we report a negative coefficient of the variable SIZE in OLS and IV estimations, and a positive one in FE estimations, but the coefficient of the variable SIZE is not always significant. Demsetz and Villalonga (2001) also reported a non-significant relation between firm size and firm performance. Summing up, we find a positive relationship between accounts receivable and firm profitability. According to our initial expectations, there are differences in the value of receivables according to firm characteristics. In this sense, we find higher profitability of receivables for larger and more liquid firms (who suffer less credit constraints). Furthermore, uncertain demand firms have higher receivables profitability. Thus, the evidence is partially consistent with, and therefore supports, the financial and operational motives for trade credit. However, we do not find results supporting the commercial motive for trade credit, since we find lower profitability of receivables for less market presence firms and for small firms without reputation in product markets. 7. CONCLUSIONS Trade credit management is particularly important in the case of small and medium-sized companies, most of whose assets are in the form of current assets. Efficient trade credit management could improve firm profitability significantly. Though the impact of trade credit policy on profitability is highly important, no studies have been carried out to examine this relationship. In this context, the objective of the current research is to provide empirical evidence about the effect of trade credit on the profitability for a sample of Spanish SMEs, due to their particular institutional setting characterized by less developed capital markets (La Porta, Lopez-de-Silanes, Shleifer, and Vishny, 1998), lack of creditor protections and weak legal system. We find a positive linear relationship between the investment in trade credit and firm profitability derived from the fact that the benefits associated to trade credit surpass the costs of vendor financing. Furthermore, the effect of receivables on firm profitability 18

19 differs depending on certain firms characteristics. According to the financial motive for trade credit, larger and more creditworthy firms will extend trade credit to their smaller customers, so increasing firm s sales and generating an implicit rate of return. In this sense, we find unconstrained firms, e.g., larger and more liquid firms, obtain higher returns on receivables compared to smaller and less liquid firms. The operational motive for trade credit predicts that firms with variable demand will extend more trade credit than firms with relatively stable demand. We find evidence consistent with the view that trade credit help firms to smooth demand, since results show higher profitability of receivables for the subsample of uncertain demand firms than for stable demand firms. Nevertheless, in a certain sense, our results are contrary to the commercial motive for trade credit. We do not find that it is more profitable for firms without an established reputation to extend trade credit, nor for less market share firms. This paper shows the importance of trade credit management in value generation in small and medium-sized firms. 19

20 REFERENCES Almeida H, Campello M, Weisbach MS, The cash flow sensitivity of cash. Journal of finance 59, Arellano M, Bond S, Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations. The review of Economics Studies 58, Atanasova C, Access to Institutional Finance and the Use of Trade Credit. Financial Management, Baltagi BH, Econometric Analysis of Panel Data, 2nd ed., John Wiley & Sons, Chichester. Berger PG, Ofek E, Diversification s effect on firm value. Journal of Financial Economics 37, Berger AN, Udell GF, The economics of small business finance: The Roles of Private Equity and Debt Markets in the Financial Growth Cycle. Journal of Banking & Finance 22, Boissay F, Gropp R, Trade credit defaults and liquidity provision by firms. European Central Bank, Working Paper Series, No Brennan MJ, Maksimovic V, Zechner J, Vendor Financing. Journal of Finance 43, Cheng NS, Pike R, The trade credit decision: evidence of UK firms. Managerial and Decision Economics 24, Cuñat V, Trade credit: suppliers as debt collectors and insurance providers. Review of Financial Studies 20, Deloof M, Does Working Capital Management affect profitability of Belgian firms?. Business Finance and Accounting 30, Deloof M, Jegers M, Trade credit, product quality, and intragroup trade: some European evidence. Financial Management 25, Deloof M, Jergers M, Trade Credit, Corporate Groups, and the Financing of Belgian Firms. Journal of Business Finance and Accounting 26, Demirguc-Kunt A, Maksimovic V, Law, Finance, and Firm Growth. The Journal of Finance 53,

21 Demirguc-Kunt A, Maksimovic V, Funding growth in bank-based and marketbased financial systems. Journal of Financial Economics 65, Demsetz H, Villalonga B, Ownership structure and corporate performance. Journal of Corporate Finance 7, Dittmar A, Mahrt-Smith J, Corporate governance and the value of cash holdings. Journal of Financial Economics Elliehausen GE, Wolken JD, The demand for trade credit: An investigation of motives for trade credit use by small businesses. Working Paper, The Federal Reserve Board. Emery GW, A pure financial explanation for trade credit. Journal of Financial and Quantitative Analysis 19, Emery GW, An optimal financial response to variable demand, Journal of Financial and Quantitative Analysis 22, Emery GW, Positive Theories of Trade Credit. Advances in Working Capital Management 1, Emery GW, Nayar N, Product quality and payment policy. Review of Quantitative Finance and Accounting 10, Faulkender and Wang, Corporate Financial Policy and the Value of Cash. Journal of Finance 61, Ferris JS, A transactions theory of trade credit use, Quarterly Journal of Economics 96, Fisman R, Love I, Trade Credit, Financial Intermediary Development, and Industry Growth. The Journal of Finance 58, Frank MZ, Maksimovic V, Trade credit, collateral, and adverse Selection. WP. Garcia-Teruel PJ, Martinez-Solano P, Determinants of trade credit: A comparative study of European SMEs. International Small Business Journal 28, Giannetti M, Do better institutions mitigate agency problems? Evidence from corporate finance choices. Journal of financial and quantitative analysis 38, Harris M, Raviv A, The Theory of Capital Structure. The Journal of Finance 46, Hernandez de Cos, Hernando I, El crédito comercial en las empresas manufactureras españolas. Moneda y Crédito 209,

22 Hill MD, Kelly GW, Lockhart GB, Washam JO, Trade Credit, Market Value, and Product Market Dynamics, FMA New York. Hsiao C, Benefits and Limitations of Panel Data. Econometric Reviews 4, Jensen M, Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76, Joh SW, Corporate governance and firm profitability: Evidence from Korea. Journal of Financial Economics 68, La Porta R, Lopez-de-Silanes F, Shleifer, A and Vishny R, Law and finance. Journal of Political Economy 106, Lang LH, Stulz RM, Tobin s q, Corporate Diversification, and Firm Performance. Journal of Political Economy 102, Lee YW, Stowe JD, Product risk, asymmetric information, and trade credit. Journal of Financial and Quantitative analysis 28, Lewellen W, McConnell J, Scott J, Capital markets influences on trade credit policies. Journal of Financial Research 1, Long MS, Malitz IB, Ravid SA, Trade Credit, Quality Guarantees, and Product Marketability. Financial Management 22, Marotta G, Is trade credit more expensive than bank loans? Evidence form Italian firm-level data. Working Paper (Universit`a di Modena e Reggio Emilia). Meltzer AH, Mercantile Credit, Monetary Policy, and Size of Firms. The Review of Economics and Statistics 42, Mian S, Smith C, Accounts receivable management policy: theory and evidence. Journal of Finance 47, Michaelas N, Chittenden F, Poutziouris P, Financial Policy and Capital Structure Choice in U.K. SMEs: Empirical Evidence from Company Panel Data. Small Business Economics 12, Modigliani F, Miller M, Corporate income taxes and the cost of capital: A correction. American Economic Review 48, Mulhern A, The SME sector in Europe: a broad perspective. Journal of Small Business management 33,

23 Myers SC, Majluf NS, Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics 13, Nadiri NI, The determinants of trade credit terms in the U.S. total manufacturing sector. Econometrica 37, Ng CK, Smith JK, Smith RL, Evidence on the determinants of credit terms used in interfirm trade. Journal of Finance 54, Niskanen J, Niskanen M, The Determinants of Corporate Trade Credit Policies in a Bank-dominated Financial Environment: the Case of Finish Small Firms. European Financial Management 12, Paul S, Boden R, The secret life of UK trade credit supply: Setting a new research agenda. The British Accounting Review 40, Peel MJ, Wilson N, Howorth C, Late Payment and Credit Management in the Small Firm Sector: Some Emprical Evidence. International Small Business Journal 18, Petersen MA, Rajan RG, The Benefits of Lending Relationships: Evidence from Small Business Data. Journal of Finance 49, Petersen MA, Rajan RG, Trade credit: theories and evidence. Review of Financial Studies 10, Pettit RR, Singer RF, Small Business Finance: A Research Agenda. Financial Management 14, Pike R, Cheng NS, Credit management: An Examination of Policy Choices, Practices and Late Payment in UK. Journal of Business Finance and Accounting 28, Pike R, Cheng NS, Cravens K, Lamminmaki D, Trade credits terms: asymmetric information and price discrimination evidence from three continents. Journal of Business, Finance and Accounting 32, Rajan RG, Zingales L, Financial dependence and growth. American Economic Review 88, San-Jose L, Cowton CJ, The Trade Credit and Credit Crunch: a Descriptive Analysis in Europe, UK and Spain. Working Paper (University of Huddersfield). Scherr FC, Hulburt H, The Debt Maturity Structure of Small Firms. Financial Management 30,

24 Schwartz RA, An economic model of trade credit. Journal of Financial and Quantitative Analysis 9, Smith JK, Trade Credit and Informational Asymmetry. Journal of Finance 42, Stulz RM, Managerial Discretion and Optimal Financing Policies. Journal of Financial Economics 26, Titman S, Wessels R, The Determinants of Capital Structure Choice. Journal of Finance 43, Wilner BS, The exploitation of relationship in financial distress: the case of trade credit. Journal of Finance 55, Wilson N, Summers B, Trade credit terms offered by small firms: survey evidence and empirical analysis. Journal of Business Finance and Accounting 29,

25 Table 1 Descriptive Statistics Variable Obs Mean Std. Dev. perc 10 Median Perc 90 ROA REC ADJUSTEDREC SIZE GROWTH DEBT GDPGR SALESVOL LIQ MKSHARE Table 2 Correlation Matrix ROA REC ADJREC SIZE GROWTH DEBT GDP SALESVOL LIQ MKTSHARE ROA REC ADJREC SIZE GROWTH DEBT GDP SALESVOL LIQ MKTSHARE

26 Table 3 Accounts receivable and ROA by industry Obs. ROA REC 1 Manufacture of food products Manufacture of beverages Manufacture of tobacco products 0 4 Manufacture of textiles Manufacture of wearing apparel Manufacture of leather and related products Manufacture of wood and of products of wood and cork, except furniture; manufacture of articles of straw and plaiting materials 8 Manufacture of paper and paper products Printing and reproduction of recorded media Manufacture of coke and refined petroleum products Manufacture of chemicals and chemical products Manufacture of basic pharmaceutical products and pharmaceutical preparations Manufacture of rubber and plastic products Manufacture of other non-metallic mineral products Manufacture of basic metals Manufacture of fabricated metal products, except machinery and equipment Manufacture of computer, electronic and optical products Manufacture of electrical equipment Manufacture of machinery and equipment n.e.c Manufacture of motor vehicles, trailers and semi-trailers Manufacture of other transport equipment Manufacture of furniture Other manufacturing Repair and installation of machinery and equipment ANOVA ANOVA is p-value of ANOVA test. It provides a statistical test of whether or not the means of several groups are all equal. If the null hypothesis is rejected, there are significant differences between groups. ROA is the return on assets; ratio of earnings, before interest and taxes to total assets. REC is the investment in trade credit; receivables to total assets. 26

27 Table 4. Mean values of ROA by REC deciles Range of REC ROA 1st decile % 2nd decile % 3rd decile % 4th decile % 5th decile % 6th decile % 7th decile % 8th decile % 9th decile % 10th decile % ROA is the return on assets; ratio of earnings, before interest and taxes to total assets. REC is the investment in trade credit; receivables to total assets. Table 5 Effect of trade credit on profitability (I) (1) (2) (3) (4) (5) (6) OLS OLS FE FE IV IV REC *** *** *** *** *** *** (35.30) (8.13) (27.28) ( 8.03) (34.44) (4.05) REC *** (1.23) ( 1.16) (3.43) SIZE *** *** *** *** *** *** ( -4.75) ( ) (21.14) (21.09) (-3.45) ( -3.12) GROWTH *** *** *** *** *** *** (66.07) (66.05) ( 60.43) (60.43) (62.47) ( 62.38) DEBT *** *** *** *** *** *** (-76.72) ( ) (-67.96) (-67.96) (-70.22) (-70.04) GDP ( -0.07) ( ) (-0.17) (-0.17) (-0.62) (-0.62) Constant * * (1.85) (1.86 ) (1.23) (1.26) (0.85) (0.86) R-squared Hausman Observations The dependent variable is ROA (Return on Assets). REC investment in trade credit (receivables to total assets); REC2 REC squared; SIZE company size; GROWTH sales growth; DEBT debt to total assets; GDP annual GDP growth. Results obtained using ordinary least squared, fixed-effects, and instrumental variables estimations. t statistics in brackets. ***significant at 1%, **significant at 5%, *significant at 10% level. Hausman is p-value of Hausman (1978) test. If null hypothesis rejected, only within-group estimation is consistent. If accepted, random-effects estimation is best option, since not only it is consistent, but it is also more efficient than the within-group estimator. Time and sectorial dummies are included in all regressions, although coefficients are not presented. 27

28 Table 6 Effect of trade credit on profitability (II) (1) (2) (3) (4) (5) (6) OLS OLS FE FE IV IV REC *** *** *** (39.13) (28.70) (34.04) ADJUSTEDREC *** *** *** (35.26) (27.13) (34.37) ARDEVIATION *** *** *** (21.17) ( 10.44) (23.22) REC ARDEVIATION *** *** (-5.34) ( 0.84) (-11.96) SIZE *** *** *** *** *** *** (-4.73) (6.28) (21.19) (25.61) (-3.44) (8.14) GROWTH *** *** *** *** *** *** (66.1) (64.32) (60.51) ( 56.99) (62.53) (59.97) DEBT *** *** *** *** *** *** (-76.71) (-77.93) (-67.97) (-66.31) (-70.22) (-71.81) GDP (-0.20) (-0.00) (-0.37) (-0.28) (-0.61) (-0.27) Constant ** * (2.18) (1.20) ( 1.80) ( 0.40) (0.87) (0.39) R-squared Hausman Observations The dependent variable is ROA (Return on Assets). REC investment in trade credit (receivables to total assets); ADJUSTEDREC is REC less industry mean REC; ARDEVIATION is a dummy variable that takes the value one if the firm grants shorter credit period than the industry mean; REC ARDEVIATION is receivables to assets ratio multiplied by ARDEVIATION; SIZE company size; GROWTH sales growth; DEBT debt to total assets; GDP annual GDP growth. Results obtained using ordinary least squared, fixed-effects, and instrumental variables estimations. t statistics in brackets. ***significant at 1%, **significant at 5%, *significant at 10% level. Hausman is p-value of Hausman (1978) test. If the null hypothesis is rejected, only within-group estimation is consistent. If accepted, random-effects estimation is best option, since not only it is consistent, but it is also more efficient than the within-group estimator. Time and sectorial dummies are included in all regressions, although coefficients are not presented. 28

29 Table 7 Firm characteristics and profitability of receivables (I) (1) (2) (3) (4) (5) (6) OLS OLS FE FE IV IV REC *** *** *** *** *** *** (27.68) (26.98) (20.47) (26.52) (28.49) (24.93) SIZE *** *** *** *** * (-6.99) (0.61) (19.15) (21.21) (-5.47) (1.70) GROWTH *** *** *** *** *** *** (66.09) (64.85) (60.42) (59.34) (62.51) (61.66) DEBT *** *** *** *** *** *** (-76.77) (-65.69) (-67.95) (-65.91) (-70.74) (-59.94) GDP (-0.07) (-0.02) (-0.16) (-0.21) (-0.54) (-0.76) DSIZE *** (0.09) (-0.94) (6.72) REC DSIZE *** *** (-4.45) (-0.57) (-11.22) DLIQ *** *** *** (-13.45) (-7.16) (-5.34) REC DLIQ *** *** *** (-2.79) (-3.68) (-7.66) Constant ** * (2.00) (1.70) (1.33) (1.25) (0.78) (0.96) R-squared Hausman Observations The dependent variable is ROA (Return on Assets). REC investment in trade credit (receivables to total assets); SIZE company size; GROWTH sales growth; DEBT debt to total assets; GDP annual GDP growth; DSIZE is a dummy variable that takes the value one whether SIZE is less than the median firm size; REC DSIZE is receivables to assets ratio multiplied by DSIZE; DLIQ is a dummy variable that takes the value one whether LIQ is less than the median firm liquidity; REC DLIQ is receivables to assets ratio multiplied by DLIQ. Results obtained using ordinary least squared, fixedeffects, and instrumental variables estimations. t statistics in brackets. ***significant at 1%, **significant at 5%, *significant at 10% level. Hausman is p-value of Hausman (1978) test. If the null hypothesis is rejected, only within-group estimation is consistent. If accepted, random-effects estimation is the best option, since not only it is consistent, but it is also more efficient than the within-group estimator. Time and sectorial dummies are included in all regressions, although coefficients are not presented. 29

30 Table 8 Firm characteristics and profitability of receivables (II) (1) (2) (3) (4) (5) (6) OLS OLS FE FE IV IV REC *** *** *** *** *** *** (29.47) (22.31) (-25.73) (19.52) (28.56) (24.01) SIZE *** *** *** *** *** *** (-4.31) (-24.47) (-20.17) (16.78) (-3.06) (-20.00) GROWTH *** *** *** *** *** *** (64.84) (65.15) (-62.15) (58.10) (61.47) (61.50) DEBT *** *** *** *** *** *** (-80.71) (-76.94) (-68.62) (-66.99) (-73.86) (-70.82) GDP (-0.14) (-0.20) (-0.19) (-0.21) (-0.70) (-0.70) DSALESVOL *** *** (-8.10) (-5.27) (-0.56) REC DSALESVOL *** *** *** (-5.42) (-3.99) (-10.79) DMKTSHARE *** *** *** (-15.49) (-11.57) (-4.82) REC DMKTSHARE ** *** (-2.03) (-0.29) (-8.90) Constant ** *** * (2.06) (2.83) (1.45) (1.90) (0.93) (1.03) R-squared Hausman Observations The dependent variable is ROA (Return on Assets). REC investment in trade credit (receivables to total assets); SIZE company size; GROWTH sales growth; DEBT debt to total assets; GDP annual GDP growth; DSALESVOL is a dummy variable that takes the value one if SALESVOL is less than the median sales volatility; REC DSALESVOL is receivables to assets ratio multiplied by DSALESVOL; DMKTSHARE is a dummy variable that takes the value one whether MKTSHARE is less than the median market share; REC DMKTSHARE is receivables to assets ratio multiplied by DMKTSHARE. Results obtained using ordinary least squared, fixed-effects, and instrumental variables estimations. t statistics in brackets. ***significant at 1%, **significant at 5%, *significant at 10% level. Hausman is p-value of Hausman (1978) test. If the null hypothesis is rejected, only within-group estimation is consistent. If accepted, random-effects estimation is best option, since not only it is consistent, but it is also more efficient than the within-group estimator. Time and sectorial dummies are included in all regressions, although coefficients are not presented. 30

31 LIST OF FIGURES Figure 1: REC-ROA relationship by industry Figure 2: Mean value of ROA for each decile of REC 31