Determinants of Corporate Liquidity for Public Firms in the Netherlands
Cash holdings versus lines of credit
August 2011
Petra Sannes (s1464957) Supervisor: Dr. H. Gonenc University of Groningen (RUG)
Keywords: Corporate Liquidity Demand, Trade-off, Credit Lines, Cash holdings JEL-Classification: E41, G3
Abstract
_____________________________________________________________________
In this paper we investigate the choice of corporate liquidity for public firms in the Netherlands. We will dive deeper into the substitutability of bank lines versus holding cash; this will be explored and compared for the period 2001-2010. Firms with high levels of cash are indifferent between credit lines and cash holdings, while this is not the case for firms with low levels of cash. Our results do not indicate a discrepancy between financial constrained and financial unconstrained firms with respect to credit lines and cash holdings.
Introduction
Why do firms choose primarily for pre-committed sources of funding to organize their future liquidity needs?
Nowadays there are different ways to fund liquidity management. Next to holding cash or using a credit line, firms may choose for future operating cash holdings or proceeds from future debt issuances. However, these operating cash holdings or proceeds from future debt issuances will expose the firm to additional risks since their availability depends directly on firm performance (Acharya and Almeida, 2009).
Demiroglu and James (2011) state that while managers might be tempted to use cash reserves opportunistically to increase their own wealth at the expense of the firm’s shareholders, the use of credit lines can overcome this managerial problem (i.e. by providing liquidity only when valuable projects arise). Moreover interest payments can be deducted from taxable income with respect to credit lines, which is not the case for cash holdings. Finally cash earns less than the debt used to fund it (Demiroglu and James, 2011).
Although the many advantages of credit lines, firms hold in general large amounts of cash.
Therefore we will also take a closer look at the costs of credit lines. One disadvantage might be that very risky firms may not be able to obtain a credit line from the bank. However Sufi (2009) and Demiroglu and James (2009) show that 87% of public firms and 64% of large private firms have access to lines of credit. Another disadvantage is that with a liquidity shortfall banks are reluctant to provide credit lines. The ability of the banking sector to meet corporate liquidity needs will depend on the extent to which firms are subject to correlated systematic liquidity shocks (Acharya and Almeida, 2009). Hence aggregate risk will create a cost of credit lines. To explore further the reasons and tradeoffs firms make with corporate liquidity management we will dive deeper into this subject.
Although there has been written much about cash holdings and lines of credit, there is an important gap in the tradeoff and substitutability of these two sources of funding. This is mainly due to the limited availability of data with respect to lines of credit.
In this paper we extend and improve existing literature on the policy of corporate liquidity for public firms. A new firm-level data set will be exploited for large non-financial firms in the Netherlands. The data set covers the ten year period 2001-2010 and therefore we are able to measure cross-sectional characteristics of liquidity holdings and to include a time dimension
as well. My primary research question is what determines the choice of corporate liquidity for public firms in the Netherlands. Common liquidity measures are the quick ratio, the current ratio and the operating cash flow ratio.
The first section will provide a literature review with the main results regarding corporate liquidity for Dutch public firms. The second section contains the methodology and the third section data description. The fourth section provides the empirical results. The fifth section offers a conclusion and the final sections present the references and appendices.
Literature review
Theoretical Background
In a world of perfect capital markets, the amount of cash holdings of a firm has no impact on its firm value (Hubensack et al. 2010). However when there are imperfect capital markets, a firm has benefits as well as costs and thus will hold an optimal level of cash (i.e. tradeoff theory of cash, Miller and Orr (1966)).
Opler et al. (1999) finds evidence confirming a static tradeoff model of cash holdings using a sample of publicly traded U.S. firms. They find that riskier firms with large growth
opportunities hold relatively high ratios of cash to total non-cash assets. Firms with good access to external markets tend to hold lower ratios of cash to total non-cash assets. This is in line with the view that firms hold more liquidity to make certain that they will be able to invest in the future (when cash relative to investments are too low and external financing too expensive). Moreover Opler et al. (1999) shows that firms that tend to do well collect more cash than predicted by the static tradeoff model in which managers maximize shareholder wealth.
Following theoretical literature (e.g. Opler et al. (1999), Hubensack et al. (2010)), holding lines of credit allows firms to diminish their cash holdings as the supplementary liquidity reduces the benefit of more cash holdings. In other words, this suggests there will be a negative relationship between lines of credit and cash holdings.
Credit Lines
Sufi (2009) and Yun (2009) find that lines of credit are enormous sources of liquidity for the firms in the United States. A line of credit is an arrangement in which a bank or vendor agrees to lent money to a customer over a particular period and at predetermined terms.
According to Sufi (2009) firms with high cash holdings make greater use of credit lines, while the opposite is true for firms with low levels of cash. He finds that firms with low levels of cash also are less likely to receive a credit line from the bank. The results are suggesting that maintenance of high cash holdings determines the choice for corporate liquidity. Furthermore, firms should maintain high levels of cash in order to remain compliant with financial
covenants. Demiroglu and James (2010) assert that firms that select tighter financial covenants are those that are riskier and have fewer investment opportunities.
Author Country Company Period Dataset Main Results
Opler, Pinkowitz, Stulz and Williamson (1999)
United States Public firms. 1971-1994 1048 firms Evidence of static tradeoff model of cash holdings.
Sufi (2009) United States Public firms. 1996-2003 4604 firms (31.533 firm- year
observations).
More use of credit lines with higher current or expected cash.
Yun (2009) United States Public firms. 1987-2000 2533
observations for 212 firms
Firms increase cash relative to lines of credit when threat of takeover weakens.
Lins, Servaes and Tufano (2010)
International sample. Largest representation Germany, U.S.
and Japan.
Both public and private firms.
2005 survey of chief financial officers
204 firms from 29 countries.
Non-operational cash as general insurance and lines of credit to fund future growth.
Demiroglu and James (2010)
United States Both public and private firms.
1995-2004 Final IPO sample consists of 1350 deals.
Access bank lending private firms
significantly more sensitive to internally generated funds and credit
market conditions.
Demiroglu and James (2011)
Review empirical evidence
Both public and private firms.
1996-2009 n/a Use of lines of
credit dependent on financial condition.
Drobetz, Grüninger and Hirschvogl (2010)
International sample.
Public firms. 1995-2005 8661 firms with 48.240 firm- year observations and 45 countries.
Value corporate cash holdings lower with higher degree information asymmetry.
Hubensack, Pfingsten and Schertler (2010)
Germany Small and medium sized firms.
October 2002 – December 2008
1525 clients. Cash substitution effects SMEs.
Flannery, Lockhart (2009)
United States Public firms. 1996-2006 Full sample 4640 firms and 26.215 observations.
Shareholders of financially- unconstrained firms value cash and credit lines similarly.
Strong evidence that transactions costs shape financial policy.
Huang (2009) United States Publicly-traded bank holding companies.
May 2007- November 2008
Final sample 120 banks.
Credit lines provided only conditional insurance for a number of borrowers during crisis since banks are likely to influence takedown volumes in short term.
Demiroglu and United States Non-financial January 1995- 7237 loans from Firms with less
James (2009) public firms December 2001 2813 borrowers. investment opportunities and riskier firms decide on tighter financial covenants.
Campello, Graham and Harvey (2010)
United States, Europe and Asia.
Public and private firms.
December 2008 1050 Chief Financial Officers (CFOs) in 39 countries.
Constrained firms withdrew their money from outstanding lines of credit.
Ivashina and Scharfstein (2009)
United States Public firms. Firms with lines contracted before July 2007 and outstanding through the first half of 2009.
544
manufacturing firms
Firms whose lines expired and were not renewed reduced their investments and increased cash holdings.
Kalcheva and Lins (2007)
United States Public firms. 1995-1999 5102 firms from 31 countries
If shareholder protection is weak, firm value is lower when managers hold more cash.
Martin and Santomero (1997)
United States Both public and private firms.
n/a n/a Demand credit
line is positively related to business growth prospects and negatively related to uncertainty in those prospects.
Nevertheless the authors state that insurance of credit lines is not complete; it is depending on both the borrower’s and the lender’s financial health. Besides risk characteristics and
investment opportunities, financial covenants are based on contracting parties’ expectations with respect to future changes in the borrowers covenant variable and the likelihood and impact of a covenant violation (Demiroglu and James, 2011). Rocco Hung (2009) also found evidence of partial insurance for (smaller, riskier and shorter-term) borrowers during the credit crunch.
According to Martin et al. (1997) the firm’s credit line demand is positively related to business growth prospects and negatively related to uncertainty in those prospects.
Furthermore, Yun (2009) find that firms increase cash compared to lines of credit when the threat of takeover declines. This trend is less seen for firms with good internal governance.
Incentive problems, transaction costs and asymmetric information may all raise external financing. In such a case the firm may want to reserve some form of liquidity for the future.
Demiroglu and James (2010) conclude that corporate cash holdings are lower in states in which there is a high level of information asymmetry.
When shareholder protection is weak, the value of the firm is lower when controlling managers hold more cash (Kalcheva and Lins, 2007). This might increase somewhat due to paying dividends. However, when shareholder protection is strong they find no relationship between holding cash by the controlling managers and firm value.
Third, the impact of credit constraints on real firm behavior plays an important role. Campbell et al. (2010) indicates that constrained firms cut more in spending (i.e. technical and capital spending, employment). Furthermore they demonstrate that constrained firms used more cash, drew more from their agreements with the bank due to fear that access would be restricted in the future, and more assets were sold in order to fund their operations.
Flannery and Lockhart (2009) find that financially-unconstrained firms value credit lines similarly to cash holdings. This does not hold for financially-constrained firms which are likely to save more cash out of their cash holdings. Their results indicate that transaction costs have a major impact on financial policy, and that shareholders benefit from low barriers of access to liquidity.
Shareholders value cash more in firms with low levels of cash holdings, low leverage and financial constraints in financial markets. According to Faulkender and Wang (2006) the
marginal value of cash ranges from 28 to 63 cents higher for the mean constrained firm-year than the mean unconstrained firm. When there will be a focus on the subset of firms that are likely to raise capital in the near future the results are even stronger. There will be an optimal level of cash when marginal costs equal marginal benefits (Miller and Orr, 1966). Marginal benefits arise when future costs of liquidity or opportunity costs are prevented, while marginal costs take place from the divergence of the rent of cash assets and alternative investments (Miller and Orr, 1966).
Finally, Hubensack et al. (2010) and Huang (2009) do not agree fully with each other with respect to the substitution of credit lines and cash holdings. While Hubensack et al. (2010) find evidence of cash substitution for the German small and medium-sized enterprises, Huang (2009) suggests that credit lines offer only partial insurance. The latter paper shows why credit lines are not a perfect substitute for cash holdings. Lins et al. (2007) looks also at the substitution between cash holdings and lines of credit. According to their survey 41% of the sampled firms view cash holdings and lines of credit as substitutes whereby large bank lines of credit bring about low cash balances and vice versa.
Hypotheses
All these authors have an own opinion on the substitution between cash holdings and lines of credit. Since there are opposing views this will be explored more in detail. According to the tradeoff theory there should be a negative relationship between cash holdings and lines of credit. On the other hand, critics such as Sufi et al. (2009) argue that firms must uphold large cash holdings to remain compliant with covenants, and banks limit firm access to credit facilities in response to covenant violations.
Hypothesis: There is a negative relationship between lines of credit and cash.
To describe and evaluate the substitution between credit lines and cash, we attempt to explain the movements in the former by reference to movements in the latter variable. We focus on the year of receiving a credit line, since that will be the moment for which the bank decides on to accept or reject any lending.
According to Faulkender and Wang (2006) additional cash is most valued by shareholders of a firm with low levels of cash, low leverage and when financial constraints are involved.
Although firms that maintain liquidity are valued higher, there is a limit to this increasing marginal value of cash. The results of Faulkender and Wang (2006) are replicated in the article of Flannery (2009).
Flannery and Lockhart (2009) classify financially constrained and unconstrained firms by focusing at payout ratio, size and existence of a bond rating. With respect to payout ratio, the results are considered as constrained (unconstrained) if the payout ratio is in the lower (upper) 30th percentile by year. For size it is considered constrained (unconstrained) if total assets are in the lower (upper) 30th percentile by year. Finally Flannery (2009) state that if a firm has a bond rating at a year-ending which it reports positive debt, the result is deemed unconstrained, but constrained if there is positive debt and no bond rating.
Hypothesis: Firms prefer cash when financial constraints are involved.
Data & Methodology
Sample
We construct a sample of firms by merging LexisNexis and DataStream with corresponding public announcements for the period 2001-2010. These data include non-financial firms which have obtained a credit line in the Netherlands. Financial firms will be excluded since they may hold cash in order to meet capital requirements. Our final data set includes 45 firms that have an agreement with the bank regarding a credit line. It is important to note that there is no data ready available and therefore has been collected manually.
Furthermore, in order to find additional data of the companies involved we use the database Orbis in which Dutch and foreign newspapers are stored. With the search strategy of Orbis it is possible to find information for listed and unlisted companies in the Netherlands in the period under study.
The hypotheses will be tested by means of panel data. Panel data have the dimensions of both time series and cross-sections.
In order to deal with the panel data we will use the method of least squares. A single equation will be estimated on all the data, so that the dataset for x is stacked up into a single column containing all the cross-sectional and time-series observations, and similarly all of the
observations on each explanatory variable would be stacked up into single columns in the y matrix. Then this equation would be estimated in the usual fashion using panel least squares.
Definition of variables
In the tests that we run there are different variables that we use. It is useful to explain them in detail.
We follow DataStream in the definitions of the variables that we use for the credit line tests.
Cash is defined as the money available for use in the normal operations of the company. It is the most liquid of all of the company’s assets (WC02003). The other variables that we examine in our tests are Cash and short term investments, which represent the sum of cash and short term investments (WC02001). Total assets represent the sum of total current assets, long term receivables, investment in unconsolidated subsidiaries, other investments, net property plant and equipment and other assets (WC02999). Total liabilities represent all short and long term obligations expected to be satisfied by the company (WC03351). EBITDA represent the earnings of a company before interest expense, income taxes and depreciation. It is calculated by taking the pretax income and adding back interest expense on debt and
depreciation, depletion and amortization and subtracting interest capitalized (WC18198).
Market value to book value is defined as the market value of the ordinary (common) equity divided by the balance sheet value of the ordinary (common) equity in the company. Please note that this item is at the security level. To derive market to book value at the company level: set up an expression: MVC/WC03501 (MTBV). Enterprise value is defined as market capitalization at fiscal yearend date + preferred stock + minority interest + total debt minus cash. Cash represents cash & due from banks for banks, cash for insurance companies and cash & short term investments for all other industries (WC18100). The Payout ratio is the ratio of dividends per share to earnings per share for the last financial period (POUT).
Dividend payout per share (%) is defined as Dividends per share / earnings per share * 100.
This item is also available at the security level for 1987 and subsequent years (WC09504 company and W09504 security). Number of employees (DWEN). Price (Adjusted – Default) represents the official closing price. This is the default data type for all equities (P). Cash /sales (%) represents the profitability ratio and is measured by Funds from Operations / Net sales or Revenues * 100 (WC08311). Net sales or revenues represent gross sales and other operating revenue less discounts, returns and allowances (WC01001). Market Capitalization represents Market Price-Year End * Common Shares Outstanding (WC08001).
Representativeness, Normal distribution, Outliers
The sample is representative with respect to large public firms in the Netherlands (Table 11).
In order to test for normality we will test for Bera-Jarque (BJ). The BJ test uses the property of a normally distributed random variable that the entire distribution is characterized by the first two moments: the mean and variance (Chris Brooks, 2008). Skewness and kurtosis are the standardized third and fourth moments of a distribution (Chris Brooks, 2008). Skewness measures the extent to which a distribution is not symmetric about it mean value and kurtosis measures how fat the tails of the distribution are (Chris Brooks, 2008). A normal distribution is not skewed and is defined to have a coefficient of kurtosis of 3 (Chris Brooks, 2008). The coefficient of excess kurtosis will be zero. Finally, a normal distribution is symmetric about its mean.
Sample of credit lines
Table 1
Company name Credit Line (amount €)
Bank Duration
(year)
KPN 2.5 milliard
Consortium of Dutch and foreign banks.
3
Ordina 120 million
ING, Rabobank, Fortis Nederland and NIBC
3
ASML 500 million
Rabobank, ING, Deutsche Bank, Royal Bank of Scotland and Commerzbank.
5
Wegener 340 million - 4
Ahold 3.1 milliard
ING, ABN AMRO, Rabobank, Goldman Sachs and JP Morgan.
1
KLM-Air France 540 million
Citibank, ING and Rabobank. Good conditions.
3
Wavin 850 million
ABN Amro, Fortis, ING and Rabobank.
5
Koninklijke Ten Cate 450 million
Consortium of 11 banks. N/a
5
Akzo Nobel 750 million
Bank consortium, Barclays, ING, JP Morgan Chase and Society Generale taking charge.
6
Pharming 3 million - -
Kendrion 123.5 million
Deutsche Bank, ING Bank and Rabobank.
5
TomTom 1.585 milliard - 5.5
Imtech 300 million
ABN Amro, ING, Rabobank, KBC, Commerzbank and LBLux.
5
Neways 25 million - 7
Wereldhave 200 million ING 5
Gasunie 750 million BNP Paribas and ING. 2
Wessanen 250 million
Rabobank and Fortis Bank.
3
Reed Elsevier 2 milliard 19 banks 3
Boskalis 650 million
Rabobank, ING, Fortis and RBS.
3-5
MacIntosh Retail 260 million
ABN AMRO, ING Bank, Rabobank International and BNP Paribas Fortis.
5
Grontmij 140 million - -
BAM 750 million - 5
Wolters Kluwer 750 million ABN Amro and ING -
(Leidse) Octoplus 2 million Fortis Bank. -
Nutreco 550 million
ABN Amro, ING and Rabobank.
5
Corio 600 million
Seven banks, two of them are ING and ABN Amro.
5
USG People 700 million
Eight banks, three of them are ABN Amro, ING and Rabobank.
5
Vopak 1 milliard
Consortium of 12 banks.
5
Heijmans 400 million Bank consortium. - -
Arcelor Mittal 17 milliard
Consortium of 26 banks.
-
Heineken 2 milliard
Bank consortium of 17 banks.
6.5
Randstad 300 million - -
Nieuwe Steen 250 million ING 2-4
Arcadis 240 million
ABN Amro, ING and Rabobank.
-
Witte Molen 0,5 million - -
Roto Smeets 12.5 million ABN AMRO -
Holland Colours 10 million Dutch bank. - -
Qurius 17.5 million - -
Koninklijke Brill 4 million - -
Tie holding 350 thousand Rabobank -
Gucci 460 million Consortium of 15 ½ 1 year, ½ 3
banks. years
Philips
3.331 milliard
Consortium of banks. 5
Accell 15.5 million ABN AMRO 5
Aegon 4.232 milliard Consortium of banks. 5
Mediq 107 million - -
Results
The correlation matrix (Table 2) measures the degree of linear association between the ratios.
Cash is the event studied and expected to change whenever the amount of credit lines is altered. Since this coefficient is positive, the variables together increase or decrease.
Table 2
Cash is assumed to be ‘stochastic’, while the other variables are assumed to have fixed (‘non- stochastic’) values in repeated samples. There is a positive trend between cash and lines of credit with a coefficient of 0.735373 (i.e. degree of linear association is 73.5373%). This is in contrast to what the trade-off theory of Opler et al. (1999) predicts. EBITDA and pay-out ratio are also positively related to cash with a value of 0.744294 and 0.313757, in contrast to market to book with a value of -0.378064.
Since regression analysis as a tool is more flexible and powerful than correlation, this test will be performed as well:
1 Exactly €3.337465 milliard
2 Exactly €4.238366 milliard.
Correlation Matrix
CASH CREDIT LINES EBITDA MARKET VALUE TO BOOKPAYOUT RATIO
CASH 1 0.735373 0.744294 -0.378064 0.313757
CREDIT LINES 0.735373 1 0.709992 -0.106045 0.183034
EBITDA 0.744294 0.709992 1 -0.148678 0.240524
MARKET VALUE TO BOOK -0.378064 -0.106045 -0.148678 1 -0.14071
PAYOUT RATIO 0.313757 0.183034 0.240524 -0.14071 1
Note: CASH, CREDIT LINES and EBITDA are scaled to net sales.
See Definition of variable s for further explanations.
Table 3
Regression Results
Variable Coefficient t-Statistic
CREDIT LINES 14.44685 3.37948
EBITDA 18.97143 3.089794
MARKET VALUE TO BOOK -0.912925 -2.978791
PAYOUT RATIO 0.080552 1.212207
C 4.046241 1.512014
R² 0.72428
Adjusted R² 0.694473
Note: We use regression analysis to estimate the coefficients to model and analyse several variables with respect to liquidity. The dependent variable is cash scaled to net sales. Credit lines and EBITDA are scaled to sales as well. P-value for credit lines is 0.0017, which means that the null-hypothesis is not rejected at 0.05 and 0.01 levels. This is also the case for EBITDA and market to book (significant at 0.05 and 0.01).
The most common goodness of fit statistic is known as R². This is the square of the correlation coefficient between y and ỹ - that is, the square of the correlation between the values of the dependent variable and the corresponding fitted values from the model (Chris Brooks, 2008). Since the correlation coefficient lies between -1 and 1, R² must lie between 0 and 1. An R² of 1 indicates a perfect fit. In Table 3 we have a R² of 0.724280, which means that the explanatory variables explain 72.4280 % of the variations in the dependent variable.
The Durbin-Watson test is a test for first order autocorrelation – i.e. it tests only for a
relationship between an error and its immediately previous value (Chris Brooks, 2008). In our case we have a Durbin-Watson statistic of 0.067942 which implies that there is evidence of positive serial correlation.
Outliers
Graph 1: Histogram Standardized Residuals
0 2 4 6 8 10 12
-10 -5 0 5 10 15 20 25 30
Series: Standardized Residuals Sample 2001 2010
Observations 42 Mean 2.41e-15 Median -2.476601 Maximum 31.55721 Minimum -12.11375 Std. Dev. 9.220262 Skewness 2.149661 Kurtosis 7.527827 Jarque-Bera 68.22444 Probability 0.000000
If the residuals are normally distributed, the histogram should be bell-shaped and the Bera- Jarque statistic would not be significant (Chris Brooks, 2008). This means that the p-value given at the bottom of the normality test screen should be bigger than 0.05 to not reject the null of normality at the 5% level (Chris Brooks, 2008). In our case, the null hypothesis for residual normality is rejected very strongly (the p-value for the Bera-Jarque test is zero to six decimal places) (Chris Brooks, 2008).
Financial Constrained Firms versus Financial Unconstrained Firms
In order to determine the group of financial constrained and unconstrained firms, the data will be split according to total assets and payout (dividends) in the year of receiving the credit line.
The lower 30th percentile is classified to financial constrained firms, whereas the other firms are classified to unconstrained firms.
Table 4
Regression Results of Constrained Firms
Variable Coefficient t-statistic
Credit Lines 15.33977 4.043299
EBITDA 59.73465 4.1429
Market Value to Book -0.6209 -4.761836
Payout Ratio -0.42559 -3.350424
C -0.22221 -0.147565
R² 0.987655
Adjusted R² 0.97778
Note: We use regression analysis to estimate the coefficients to model and analyse several variables with respect to liquidity. The dependent variable is cash scaled to net sales. Credit lines and EBITDA are scaled to sales as well. P-value for credit lines is 0.0099, which means that the null-hypothesis is not rejected at 0.05 and 0.01 level. This is also the case for EBITDA and market to book (significant at 0.05 and 0.01). Pay-out ratio is significant at 0.05.
In Table 4 we have a R² of 0.987655, which means that the explanatory variables explain 98.7655% of the variations in the dependent variable. This indicates almost a perfect fit.
The Durbin-Watson statistic of 0.187307 implies evidence of positive serial correlation.
Graph 2: Histogram Standardized Residuals Constrained Firms
0 1 2 3
-4 -3 -2 -1 0 1 2 3 4
Series: Standardized Residuals Sample 2002 2010
Observations 10 Mean 1.80e-15 Median -0.080514 Maximum 3.786723 Minimum -3.443223 Std. Dev. 2.291537 Skewness 0.302422 Kurtosis 2.189053 Jarque-Bera 0.426446 Probability 0.807976
The p-value should be bigger than 0.05 to not reject the null of normality at the 5% level. This is the case with a value of 0.807976. The Bera-Jarque statistic is not significant.
Table 5
The correlation between cash is positive for all the variables except for market value to book.
Cash and credit lines have a coefficient of 0.849583 which shows a strong positive relationship.
Table 6
Regression Results Unconstrained
Firms
Variable Coefficient t-statistic
Credit Lines 13.46499 2.407867
EBITDA 18.83275 2.55225
Market Value to Book -0.97928 -0.659985
Payout Ratio 0.0216 0.208688
C 7.413909 1.395371
R² 0.660834
Adjusted R² 0.608655
Note: We use regression analysis to estimate the coefficients to model and analyse several variables with respect to liquidity. The dependent variable is cash scaled to net sales. Credit lines and EBITDA are scaled to sales as well. P-value for credit lines is 0.0234, which means that the null-hypothesis is not rejected at 0.05 level. This is also the case for EBITDA.
(Significant at 0.05). Market to book and pay-out ratio are not significant.
In Table 6 we have a R² of 0.660834, which means that the explanatory variables explain 66.0834% of the variations in the dependent variable.
The Durbin-Watson statistic of 0.024057 implies that there is evidence of a small positive serial correlation.
Correlation Matrix Constrained Firms
Cash Credit Lines EBITDA
Market to Book Payout Ratio
Cash 1 0.849583 0.936651 -0.619479 0.676789
Credit Lines 0.849583 1 0.899866 -0.198067 0.878086
EBITDA 0.936651 0.899866 1 -0.420887 0.833948
Market Value to Book -0.619479 -0.198067 -0.420887 1 -0.135862
Payout Ratio 0.676789 0.878086 0.833948 -0.135862 1
Note: CASH, CREDIT LINES and EBITDA are scaled to net sales.
See Definition of variable for further explanations.
Graph 3: Histogram Standardized Residuals Unconstrained Firms
0 2 4 6 8 10 12 14
-15 -10 -5 0 5 10 15 20 25 30 35
Series: Standardized Residuals Sample 2001 2010
Observations 31 Mean 3.32e-15 Median -3.823152 Maximum 33.34449 Minimum -11.70232 Std. Dev. 10.24209 Skewness 2.131442 Kurtosis 6.896285 Jarque-Bera 43.08124 Probability 0.000000
The p-value should be bigger than 0.05 to not reject the null of normality at the 5% level. This is not the case with a value of 0.000000. The null-hypothesis for residual normality is rejected very strongly.
Table 7
Again the independent variables all have a positive correlation with cash except for market value to book. The coefficient of cash and credit lines is 0.757234.
Table 8
Descriptive Statistics
Mean Median Mean Median Mean Median
Cash 11.59571 5.825000 14.87194 7.390000 2.362727 1.390000
Credit Lines 0.335159 0.161193 0.362334 0.180060 0.258573 0.129923 Note: The first column represents the variables (i.e. cash and credit lines). The two columns after represent the mean and median of the whole sample. Respectively the mean and median of the unconstrained firms and constrained firms follow.
Correlation Matrix Unconstrained Firms
Cash Credit Lines EBITDA Market to Book Payout Ratio
Cash 1 0.757234 0.748876 -0.222718 0.15079
Credit Lines 0.757234 1 0.732314 -0.23579 0.178627
EBITDA 0.748876 0.732314 1 -0.108256 0.127189
Market Value to Book -0.222718 -0.23579 -0.108256 1 0.040412
Payout Ratio 0.15079 0.178627 0.127189 0.040412 1
Note: CASH, CREDIT LINES and EBITDA are scaled to net sales.
See Definition of variable for further explanations.
Conclusion
One of our main findings is a positive relationship between cash holdings and lines of credit.
This might be due to the fact that banks prefer to borrow money primarily to firms which are financially healthy and are able to pay back their debts. The firms under investigation are generally large public firms trading on the Amsterdam exchange. Credit lines and cash
holdings are viewed as good substitutes when the firm owns large amounts of cash. This is not the case for firms with lower amounts of cash holdings.
Second, we find no evidence that constrained firm’s value cash more than unconstrained firms. However due to limited data this result may be biased. Although lines of credit are often known for the companies, there is still a lack of data for the years under study and important variables involved. In order to find relationships we use panel data and run regression analysis.
Our results are consistent with Lins et al. (2007) and Sufi (2009). Sufi (2009) is the first author who examined the factors that carry out the use of credit lines and cash. His dataset includes end of fiscal year information with respect to bank lines from Compustat universe for the 1996-2003 period. He finds that larger firms, firms with higher cash holdings and those with fewer investment opportunities are more likely to have access to lines of credit and therefore rely more on this source of funding (Demoriglu and James, 2011).
In this paper the determinants of corporate liquidity are investigated for public firms in the Netherlands. We contribute to the scarce empirical evidence of large non-financial firms by showing there is partial substitution between lines of credit and cash holdings. The important characteristic of our data is its consistent availability over a period of ten years.
A fruitful area of future research is more empirical analysis of private firms. Currently there is a wide gap since data is lacking for the years under study. Furthermore, although the data have to be found manually it is still not possible to find all the relevant data.
References
Acharya, V.V., H. Almeida, M. Campello (2010). “Aggregate Risk and the Choice between Cash and Lines of Credit”, NBER Working Paper Series.
Berger, A.N., G.F. Udell (1995). “Relationship Lending and Lines of Credit in Small Firm Finance”, The Journal of Business, 68(3), 351-381.
Bruinshoofd, A., C. Kool (2002). “The Determinants of Corporate Liquidity in the Netherlands”, University of Maastricht (The Netherlands).
Brooks, C. (2008). “Introductory econometrics for finance”, Cambridge University press.
Campello, M., J. Graham, C. Harvey (2010). “The real effects of credit constraints: evidence from a financial crisis.” Journal of Financial Economics, 97(3), 470-487.
Demiroglu, C., C. James (2009). “Credit Market Conditions and the Determinants and Value of Bank Lines of Credit for Private Firms.” Working Paper, Koc University.
Demiroglu, C., C. James (2010). “The information content of bank loan covenants.” Review of Financial Studies, 23(10), 3700-3737.
Demiroglu, C., C. James (2011). “The use of bank lines of credit in corporate liquidity management: A review of empirical evidence”, The Journal of Banking & Finance, 35, 775- 782.
Faulkender, M., R. Wang (2006). “Corporate financial policy and the value of cash.” Journal of Finance, 61(4), 1957-1990.
Flannery, M., B. Lockhart (2009). “Credit Lines and the Substitutability of Cash and Debt.”
Working Paper. University of Florida.
Huang, R. (2009). “How Committed are Bank Lines of Credit? Evidence from the Subprime Mortgage Crisis.” Working Paper. Federal Reserve Bank of Philadelphia.
Hubensack, C., A. Pfingsten, A. Schertler (2010). “Recallable Bank Lines of Credit and Cash Substitution – Evidence for German SMEs, University of Münster (Germany).
Ivashina, V., D. Scharfstein (2009). “Liquidity Management in the Financial Crisis.” Working Paper. Harvard Business School.
Kalcheva, I., K. Lins (2007). “International evidence on cash holdings and expected managerial agency problems.” Review of Financial Studies 20, 1087-1112.
Lins, K.V., H. Servaes, P. Tufano (2010). “What drives corporate liquidity? An international survey of cash holdings and lines of credit”, Journal of Financial Economics, 98(1), 160-176.
Martin, J.S., A. Santomero (1997). “Investment opportunities and corporate demand for lines of credit.” Journal of Banking and Finance 21, 1331-1350.
Miller, M., D. Orr (1966). “A model of the demand for money by firms”, Quarterly Journal of Economics, 80(3), 413-435.
Opler, T., L. Pinkowitz, R. Stulz, R. Williamson (1999). “The determinants and implications of corporate cash holdings”, Journal of Financial Economics, 52(1), 3-46.
Sufi, A. (2009). “Bank Lines of Credit in Corporate Finance: An Empirical Analysis”, Review of Financial Studies, 22(3), 1057-1088.
Tirole, J. (2006). “The Theory of Corporate Finance”, Princeton University Press.
Yun, H. (2009). “The Choice of Corporate Liquidity and Corporate Governance”, Review of Financial Studies, 22(4), 1447-1475.
Appendix
Table 9: Announcements Public Firms
COMPANY NAME
Announcement Public Firms
Price before Announcement (1 day)
Price after Announcement (1 day)
KPN
December 2, 2001 5.1 4.95
Ordina
October 16, 2009 4.747 4.782
ASML
June 3, 2010 23.48 23.935
Wegener
March 12, 2003 4.19 4.3
Ahold
February 24, 2003 8.1974 2.7606
KLM-Air France
July 28, 2009 15.25 15.25
Wavin
October 19, 2006 35.9496 36.5037
Koninklijke Ten Cate
December 10, 2010 28.505 28.825
Akzo Nobel
March 18, 2009 28.335 29.435
Pharming
September 26, 2002 1.08 1.13
Kendrion
December 8, 2010 14.35 14.935
Tom Tom
July 23, 2007 33.8698 37.8949
Imtech
July 18, 2007 22.68
23.0867
Neways
December 19, 2003 2.3 2.3
Wereldhave
July 1, 2002 56.05 58.5
Aegon
September 20, 2005 11.77 11.58
Wessanen
May 3, 2004 11.73 11.8
Reed Elsevier
February 17, 2009 8.821 8.93
Boskalis December 15, 2009 25.2489 25.8893
MacIntosh
June 14, 2010 16 15.8
Grontmij
May 19, 2010 15.38 15.8
BAM
August 6, 2007 17.0343 16.9552
Wolters Kluwer
June 6, 2004 14.28 14.29
(Leidse) Octoplus
July 11, 2008 1.19 1.1
Nutreco
May 20, 2009 29.445
30.55
Corio
April 6, 2005 40.4594 40.7056
USG (People)
June 14, 2007 30.4326 30.142
Vopak
August 2, 2007 21.55 21.455
Heijmans
April 29, 2009 14.5385 14.6186
Arcelor Mittal
December 6, 2006 29.6305 30.894
Heineken
December 2, 2005 23.5853 23.6142
Randstad
June 18, 2003 10.5 10.4
Nieuwe Steen
April 27, 2010 15.36 15.05
Arcadis
April 10, 2006 12.3733 12.1367
Witte Molen
June 11, 2008 10.79 10.79
Gucci
December 2, 2003 71.0979 70.2195
Holland Colours
July 2010 - -
Qurius
April 1, 2009 0.165 0.208
Roto Smeets
February 18, 2010 10.01 10
Koninklijke Brill
2006 - -
Tie Holding
October 14, 2009 0.184 0.191
Philips
May 28, 2002 25.5018 26.7827
Accell
2002 - -
Gasunie March 03, 2005 - -
Mediq
2010 - -
Table 10
Market Capitalization
Period Market Cap
2010 492866000000
2009 389835000000
2008 279059000000
2007 654129000000
2006 591317000000
Netherlands – Euronext Amsterdam N.V. – Market Capitalization of listed companies (domestic equities and exclusive foreign listings). Equity excluding investment fund shares. Euro (Securities exchange – Trading Statistics).
Table 11
Market Capitalization*
Company Market Cap
KPN 17019641
Ordina 184287
ASML 12617537
Wegener 252950
Ahold 11309452
KLM-Air France na
Wavin 577366
Koninklijke Ten Cate 701922
Akzo Nobel 10855663
Pharming 90306
Kendrion 163230
TomTom 1750066
Imtech 2480544
Neways 85809
Wereldhave 1567029
Gasunie na
Wessanen 221464
Reed Elsevier 7027239
Boskalis 3604781
MacIntosh Retail 423227
Grontmij 360285
BAM 1065428
Wolters Kluwer 4897211
(Leidse) Octoplus 46673
Nutreco 1985549
Corio 4369507
USG People 1181077
Vopak 4495651
Heijmans 253605
Arcelor Mittal 43948180
Heineken 21073722
Randstad 6716926
Nieuwe Steen 648434
Arcadis 1150540
Witte Molen 4022
Roto Smeets 44419
Holland Colours 17633
Qurius 32781
Koninklijke Brill 24668
Tie holding 9535
Gucci na
Philips 21693918
Accell 389510
Aegon 7810333
Mediq 835044
SUM 193987164
*More than 40% of total market capitalization.
Petra Sannes
Duindoornstraat 579 9741 PW Groningen
ABN AMRO
Gustav Mahlerlaan 10 1082 PP Amsterdam
Groningen, 3 maart 2011
Betreft: Scriptie MSc Business Administration – Finance
Geachte heer/mevrouw,
Voor mijn scriptie doe ik een onderzoek naar het beleid van (particuliere) ondernemingen met betrekking tot bankkrediet en contant geld. Om dit te bepalen heb ik informatie nodig van bedrijven die al dan niet een afspraak hebben met de bank (credit lines). De periode van mijn onderzoek is 2001-2010.
In afwachting op uw antwoord en, Met vriendelijke groet,
Petra Sannes
