Why New Zealand Companies Hold Cash: An Empirical Analysis

Contents

Chapter

1 Introduction

2 Theory of Corporate Cash Holdings
2.1 Static Trade-Off Model
2.2 Financing Hierarchy Model
2.3 Free Cash Flow Model
2.4 Additional Factors Influencing Corporate Cash Holdings
2.5 Summary of Theoretical Models

3 Review of Empirical Studies

4 Data
4.1 Data Description
4.2 Variable Definitions
4.3 Descriptive Statistics

5 Statistical Background
5.1 Introduction to Panel Data Models
5.2 Regression Methodologies
5.2.1 Fama-MacBeth Model
5.2.2 Cross-Sectional Regression
5.2.3 Time-Series Cross-Sectional Regression with Year Dummies
5.2.4 Fixed-Effects Regression
5.3 White Standard Errors

6 Empirical Investigations
6.1 Univariate Tests
6.2 Determinants of Corporate Cash Holdings
6.2.1 Basic Regressions on the Whole Data Sample
6.2.2 Regression with Cash Flow Variability and 1/Z-Score
6.2.3 Regression with Cash Flows from Cash Flow Statements
6.2.4 Regression Tests for Managerial Discretion
6.3 Investigation of Mean Reverting Characteristics

7 The Use of Excess Cash
7.1 Review of Current Literature on Excess Cash Holdings
7.2 Empirical Investigation of Excess Cash Holdings

8 Conclusion

List of References

A Summary of Empirical Studies

B SPSS Syntax Files
B.1 Syntax for Performing Regressions with White’s Standard Errors
B.2 Syntax for Performing a Series of Regressions

C Additional Regressions

List of Tables

2.1 Summary of model predictions

4.1 Variable definitions and predicted influence

4.2 Descriptive statistics of regression variables

6.1 Firm characteristics by cash/assets quartiles

6.2 Regression predicting firms’ cash holdings

6.3 Reduced form regression

6.4 Regression without observations with zero cash holdings

6.5 Regression with cash flow variability and 1/Z-score

6.6 Regression with cash flows from cash flow statements

6.7 Target adjustment characteristics of cash holdings

7.1 Spending based on previous year’s excess cash

7.2 Change in excess cash by decile

A.1 Overview of empirical investigations on corporate cash holdings

A.2 Overview of empirical investigations on corporate cash holdings

A.3 Overview of empirical investigations on corporate cash holdings

C.1 Transformed regression without observations with zero cash holdings

C.2 Regression without the top and bottom 10% of cash holdings

List of Figures

2.1 Optimal holdings of liquid assets

4.1 Mean cash to net assets ratio by years

5.1 Consequences of introducing dummy variables

5.2 Heteroskedasticity in an univariant linear regression

6.1 Distribution of regression coefficients

Chapter 1 Introduction

As Ross, Westerfield and Jordan (2000) report, at the end of 1997 a number of US companies piled up tremendous amounts of cash and marketable securities. At that time Ford held US $20.8 billion, GM US $14.5 billion, and Chrysler US $7.1 billion in liquid assets. This naturally raises the question for which reasons a firm would hold such large amounts of cash, and if there is an optimal amount of cash holdings.

While this issue remained largely unexamined in the financial literature for years, recently, some researchers tried to explore the determinants of corporate cash holdings in the US (e.g. Opler, Pinkowitz, Stulz and Williamson, 1999) and in some European countries (e.g. Ferreira and Vilela, 2002). However, different reasons for holding cash might exist in countries with less developed economies. In this study, I examine the determinants of corporate cash holdings using a sample of companies listed at the New Zealand Stock Exchange in the 1980-2003 period.

This work is guided by three theoretical models which are helpful in finding the de­termining factors of cash holdings. However, these models partially disagree about how certain firm characteristics affect liquid asset holdings.

The trade-off model assumes that companies weight the marginal benefits and the marginal costs of holding cash to decide on their cash holdings. Thus, firms have a target cash level to which they continuously try to adjust. Benefits of holding cash are for example the reduction of transaction costs, and costs related to holding cash are mainly opportunity costs.

The second model of corporate cash holdings is the financing hierarchy model. Ac­cording to this theory, firms want to avoid costs induced by asymmetric information and, therefore, use cash from retained earnings to finance investments before using debt or even equity.

Finally, the free cash flow theory suggests that managers have an incentive to accumu­late cash in order to increase their power and managerial discretion. Managers are able to pursue their own objectives if they can finance investments using cash and do not need to access the capital markets.

This study adds to existing analyses in the following ways. First, I examine the determinants of corporate cash holdings in an economy with less developed capital markets and a reduced availability of data. To my knowledge this has hardly been done before. Second, I combine the most promising approaches of recent studies and extent these approaches by using new methods and variables to explain corporate cash holdings.

The results of my investigation widely comply with the findings of other researchers. The main determinants of corporate cash holdings in New Zealand are a firm's growth opportunities, the variability of its cash flows, leverage, dividend payments, and the avai­lability of liquid asset substitutes. While growth opportunities and the variability of cash flows are positively related to cash holdings, large dividend payments and liquid asset substitutes indicate lower cash holdings.

The remainder of this study is organised as follows. In Chapter 2 the theoretical models of corporate cash holdings are discussed in more detail and compared regarding the predicted influences of certain firm characteristics. Chapter 3 provides an extensive review of recent studies on corporate cash holdings. Chapter 4 discusses the data basis used throughout this study and defines the variables which are used in the empirical examinations. In Chapter 5 the statistical tools used to perform the regression analyses are explained. The empirical analyses, to determine the influencing factors of corporate cash holdings, are carried out in Chapter 6. Furthermore, in this chapter the results of the analyses are discussed under consideration of the theoretical models. Chapter 7 adds to the results by providing information on the use of excess cash in New Zealand. Finally, the findings are summarized and ideas for future work are provided in Chapter 8.

Chapter 2 Theory of Corporate Cash Holdings

Assuming a world of perfect capital markets, the amount of a company’s cash holdings does not influence the wealth of its shareholders. Cash holdings are irrelevant because in case of a cash shortage a company could raise cash at zero cost. However, there are numerous capital market imperfections like transaction costs, taxes, financial distress costs, asymmetric information, and agency costs, accounting for the fact that the amount of cash holdings makes a difference for shareholders (see Soenen, 2003).

In the well-known book ”The General Theory of Employment, Interest and Money”, Keynes (1936) found three motives for companies to hold cash:

Transactions motive The company needs cash for current transactions for personal and business exchanges.

Precautionary motive The company needs cash to secure future cash needs because of the unpredictability of future cash flows.

Speculative motive The company needs cash to take advantage of possible future in­vestment opportunities.

Since Keynes (1936) published these motives for corporate cash holdings, some more advanced models have been developed. In this section the influencing factors for cash holdings according to three different models, the static trade-off model, the financing hierarchy model, and the free cash flow model are discussed.

2.1 Static Trade-Off Model

According to the static trade-off model of corporate cash holdings, the amount of cash held by a firm is determined by weighting the marginal costs and the marginal benefits of holding liquid assets (see Ferreira and Vilela, 2002).

Using the assumptions that the only objective of managers is to maximize shareholder wealth, the only cost associated with holding cash is the lower return earned on it. These opportunity costs arise from the fact that the amount of cash held could otherwise be invested at a higher rate of return. Relaxing the assumption of shareholder wealth maxi­mization as the only objective of managers, results in additional costs caused by corporate cash holdings. Managerial discretion allows for wasteful spending and acquisitions when large amounts of cash are under an manager’s control (see Dittmar, Mahrt-Smith and Servaes, 2003).

There are as well several benefits of holding liquid assets which counterbalance the costs associated with holding cash. First, cash helps to avoid transaction costs associated with the liquidation of existing assets or the raising of external funds. Second, cash enables a company to pursue the optimal investment policy and, therefore, prevents a company from rejecting positive NPV projects if external financing constraints are met. Finally, the danger of encountering financial distress is reduced by cash. If there are unexpected losses, a cash reserve acts as a buffer, especially, if external funding is difficult to obtain.

Considering the fact that buying and selling financial and real assets generates costs, an optimal amount of cash holdings exists. Transaction costs typically consist of fixed costs and variable costs, depending proportionally on the amount of cash raised. Generally, this is valid, regardless of the source of cash that is used. As a consequence, every firm has an optimal amount of liquid assets and cash cannot be viewed just as negative debt (see Opler et al., 1999; Ross et al., 2000).

The optimal amount of cash for a firm is determined by the marginal cost of a liquid asset shortage and the marginal cost of holding liquid assets. The marginal cost of cash holdings is constant because there is no evidence that the lower return of cash compared to other investments changes as the amount of cash held changes. The incremental cost of liquid asset shortage is a downward sloping curve because of the share of fixed costs to be paid for raising cash. The more liquid assets a firm is holding, the less frequently it needs to access capital markets. This relation is shown in Figure 2.1.

An increase in the costs of being short of liquid assets or an increase in the probability of being short of liquid assets would hereby shift the cost curve to the right, resulting in

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.1: Optimal holdings of liquid assets (see Opler et al., 1999)

an higher optimal amount of cash holdings (see Opler et al., 1999).

Furthermore, a precautionary and a speculative motive for holding cash exist. Ex­tending the analysis above to allow for information asymmetries and agency costs has a number of consequences. Information asymmetries between managers and investors might reduce a manager’s ability to raise outside funds and, therefore, the possibility to take advantage of all possible NPV-positive investments. Because managers know more, po­tential investors discount the value of securities to make sure they are not overpriced (see Myers and Majluf, 1984). Therefore, it is advantageous to hold cash for future investment needs in order to avoid the costs of information asymmetries of capital markets.

The assumptions of the static trade-off model, which are mentioned above, result in some financial and non-financial variables, affecting the optimal amount of liquid assets (see Dittmar et al., 2003; Ferreira and Vilela, 2002; Opler et al., 1999).

Firms with better investment opportunities are expected to hold more cash because the opportunity costs of giving up NPV-positive projects, which is the consequence of a failure in raising cash, are higher. When a firm holds large amounts of liquid assets, it can carry out projects, even if potential debtors or shareholders are not willing to finance these projects.

Large firms usually have better access to capital markets than smaller firms. Because of the substantial amount of fixed costs involved in raising external funds, it is more ex­pensive for small firms to use the capital markets. In their studies of historical importance on the demand for money by firms Baumol (1952) and Miller and Orr (1966) argue that economies of scale exist in cash management.

Firms possessing liquid asset substitutes are expected to hold less cash than other firms. In case of a cash shortage, liquid asset substitutes can be transferred to cash within a short time and at low costs. This characteristic makes it unnecessary for firms with a large amount of liquid asset substitutes to hold large amounts of cash.

According to Ferreira and Vilela (2002), the relation between leverage and the amount of liquid assets of a firm is not clear in the trade-off model of corporate cash holdings. They argue that a firm's leverage can be interpreted in two ways with respect to the trade-off model. Due to the pressure induced by amortization plans, leverage increases the risk of a firm to face financial distress in the future. To reduce the risk of bankruptcy, high leveraged firms are expected to possess more liquid assets. On the other hand, Ferreira and Vilela (2002) state that leverage can be interpreted as a proxy for a firm's ability to raise debt. Firms which have good access to the capital market are expected to hold less cash. I think that this is misleading because the current level of leverage expresses only a firm's former ability to access capital markets and not the ability to raise cash in the future. In the future a high leveraged firm will have more difficulties in accessing capital markets because a high leverage will make this more expensive. The amount of cash holding, is determined by the future ability to raise money. However, following the reasoning of Baskin (1987), debt is more expensive when a firm already has a high leverage. Because debt can be paid off using cash, a high leverage increases the opportunity costs of holding cash. Overall, the leverage ratio has no clear prediction for cash holdings in the static trade-off model.

Dividend payments are negatively related to cash holdings because firms paying di­vidends have the possibility to reduce these payments in order to raise additional funds. Firms not paying dividends need to access the capital markets instead.

While the static trade-off model implies a negative relation between cash flow and cash holdings, the relation between cash flow uncertainty and cash holdings is positive. Operating cash flows are a source of liquidity for companies. The higher the operating cash flows, the smaller is the possibility of financial distress and the less necessary is a cash reserve to ensure the capability to meet future financing demands. A high variability of these cash flows, however, increases the likelihood of financial distress and is for this reason positively related to cash holdings. Firms with a high cash flow variability cannot rely on the incoming cash flow from operations and, therefore, are unsure about the capability to meet future cash demands with the money from operating cash flow (see Dittmar et al., 2003; Ferreira and Vilela, 2002).

Another variable, which might be important for determining the optimal amount of cash holdings for a company, is the maturity of a company’s debt. Ferreira and Vilela (2002) find no clear prediction for the relation between debt maturity and cash holdings. On the one hand, they argue that firms using mostly short-term debt need to renegotiate their contracts more often and, therefore, might face the problem of financial distress if credit lines are for any reasons not renewed. This view is supported by Guney, Ozkan and Ozkan (2003), who mention that previous studies have shown that firms with high information asymmetries are likely to issue short-term debt. If short-term debt is seen as a proxy for information asymmetries, firms with a lot of short-term debt are expected to hold more liquid assets. However, Ferreira and Vilela (2002) state as well that there is empirical evidence that firms with the highest and lowest credit risk are more likely to issue short-term debt, while intermediate firms tend to issue long-term debt. Because firms with better credit ratings have better access to capital markets, they conclude that these firms will hold less cash for precautionary motives. In this paper, I follow the assumption of Ferreira and Vilela, saying that the influence of debt maturity on cash holdings cannot be clearly determined.

Due to high information asymmetries for research and development (R&D) projects, R&D expenses should be positively related to cash holdings. This is the fact because firms with high R&D expenses are more opaque, and financial distress costs are large for R&D projects (see Dittmar et al., 2003; Opler et al., 1999). However, Kim, Maurer and Sherman (1998) for example interpret R&D expenditures as an alternate proxy for a firm’s growth opportunities which (Opler et al., 1999) captures by using the market- to-book value. Thus, even if a statistical significant relation between R&D expenditures and a firm’s cash holdings is identified, it might be difficult to draw the right conclusions from this relationship.

2.2 Financing Hierarchy Model

The financing hierarchy model is based on the pecking order theory of Myers (see Myers, 1984; Myers and Majluf, 1984) which suggests that firms generally prefer internal to external financing. If it is necessary to use external financing, debt is the preferred choice to equity. Companies chose this financing hierarchy for raising cash in order to minimize costs induced by information asymmetries and other costs involved with external financing. In this model, companies are not assumed to have target cash levels, but use cash only as a necessary buffer between retained earnings and investments. Myers states that issue costs for external financing cannot be the main reason for the preferred use of internal financing because these costs are not large enough to compensate for the advantages of debt and a high leverage mentioned in the trade-off theory. However, there are costs involved with information asymmetries between the management and the shareholders. The managers know more about the true value of a project for which money should be raised than potential investors. In the pecking order model by Myers (1984), management is concerned only about the intrinsic value of the existing shares and not about the new shareholders. New investors will typically know about this behaviour of the management and, therefore, adjust the price they are willing to pay. If a firm chooses to issue stock with a market value of N, the managers know about the real value of the stock Ni, which depends on the real NPV of the project to finance. Managers will only issue stock if the real NPV of the investment opportunity, y, is larger than the difference AN = N1 — N between the market value of the stock and the real value of the investment opportunity:

Thus, a manager will always issue stock if AN is negative. In case the inside infor­mation is valuable, the management will rather choose to to pass up even an investment opportunity with positive NPV rather than issue underpriced shares. Using this rationale, a management’s announcement of new stock issues reflects negative inside information. Because investors know about this behaviour, they will adjust the price they are willing to pay for the stock, introducing additional costs of information asymmetries. Therefore, it is cheaper to use internal than external financing. A similar reasoning accounts for the preferred choice of debt to equity when external financing is needed. In case of a NPV- positive project, AN is typically smaller for debt then for equity. If a company could issue default-risk free debt, AN would be zero because there is no risk for new investors involved. But even when default-risk is considered, AN is still smaller for debt than for equity. Thus, a positive NPV project leaves more value to the old shareholders if financed by debt. Using this reasoning, a firm should try to issue equity for negative NPV projects because the new shares should be overpriced, favouring the old shareholders. However because new investors know about this behaviour, a firm would hardly be able to issue these shares, unless it already exceeded its debt capacity.

Ferreira and Vilela (2002), Opler et al. (1999), and Dittmar et al. (2003) summarize the influence of the pecking order theory on corporate cash holdings as follows:

Investment opportunities Large investment opportunities demand for a large stock of cash to pay for these projects in the most favourable way. As explained before, a company tries to avoid the necessity of using external financing, and, therefore, holds larger amounts of cash if good investment opportunities exist.

Firm size Large firms have usually been more successful in their past business operations and should have accumulated more cash after controlling for investments.

Leverage The pecking order theory implies that debt increases when investments exceed the retained earning, because a firm cannot finance its investments with excess cash. For cash holdings the opposite is accurate. If investments exceed the retained earning, cash will be used in the first place to make the investments. This implies a negative relation between debt and cash holdings.

Cash flow Controlling for the influence of other variables, firms with larger cash flows should have more liquid assets than firms with low cash flows. This is due to the fact that firms with high cash flows pay dividends, pay off debt, and accumulate cash to finance future investments. The same accounts in the opposite way for firms with low cash flows.

Dividend payments Because firms use cash flow from operations to pay dividends, pay off debt, and accumulate cash, there is a negative relation between dividend payments and cash holdings. The more dividends a company pays the less cash can be accumulated.

R&D expenses In the financing hierarchy model, firms with higher R&D expenses are expected to hold less liquid assets. Firms use their cash holdings to finance R&D projects.

However, despite the theoretical distinction between the static trade-off model and the financing hierarchy model, the difference in practice is not always as clear as it might seem. The major difference between the two models is that no optimal cash level exists in the financing hierarchy model. Cash is simply seen as negative debt and is just an outcome of financing decisions of a firm which are guided by the pecking order theory. Because in the financing hierarchy model no costs are associated with holding cash, there is no reason for shareholders or management to object against large cash holdings (Dittmar et al., 2003; Opler et al., 1999).

2.3 Free Cash Flow Model

A third theory of corporate cash holdings is the free cash flow theory of Jensen (1986). This theory partially adds to the previous models and can, to some extent, be incorporated in them. In this theory, agency costs are believed to explain the amount of a firm’s liquid assets^[1]. Jensen states that managers have a high interest in holding cash because it reduces the firm risk and increases their discretion. Even though shareholders would want excess cash to be paid out^[2], managers will hold cash to pursue their own objectives, taking negative effects for shareholder wealth into account. Liquid assets allow the management to invest in projects for which they would not get funds from the capital markets. Pay-outs to shareholders reduce the assets under a manager’s control and increase the monitoring of the capital markets because managers will need to raise new capital at a later point of time. Therefore, the extent of managerial discretion is positively related to the amount of a corporate’s liquid assets (see Bruinshoofd and Kool, 2002; Opler et al., 1999).

According to Ferreira and Vilela (2002), the free cash flow theory implies that the investment opportunities are negatively related to the amount of cash holdings. Managers with poor investment opportunities are likely to hold more cash in order to ensure the availability of funds, even for investment programs with a negative NPV. Managers would not be able to externally finance projects with poor prospects because, contrarily to financing an investment by the available cash, external financing induces monitoring from the capital markets. The monitoring of capital markets is even stronger for firms with a high leverage and, hence, firms with low leverage allow for an increased managerial discretion. The consequence of this circumstance is that managers of firms with a low leverage tend to hold a larger amount of liquid assets.

Moreover, the free cash flow theory predicts that the size of a firm and cash holdings are positively related. The shareholders of large firms are usually more dispersed than the shareholders of small firms, which supports the interest of managers for discretion. When a firm’s shareholders are highly dispersed, they are less likely to ally against mana­gers. Because of the larger amount of money needed, managers of large firm are as well less affected by market discipline through potential takeovers. Firm size and sharehol­der dispersion are takeover deterrents which result in an increased managerial discretion and power and, therefore, in increased cash holdings. The same reasoning accounts for companies with anti-takeover amendments (see Opler et al., 1999).

However, concerning the protection against takeovers, cash holdings have a negative impact, too. Even though management is independent from the capital markets and can pursue its own investment policies, cash holdings might attract potential takeovers because the takeover could be fully or partially financed using the cash available (see Opler et al., 1999). But in an empirical study Pinkowitz (2000) does not find any evidence for this relation.

Another variable influencing cash holdings according to the free cash flow theory is inside ownership. Inside ownership commonly serves as a proxy for the degree of agency conflicts found in a corporation (see e.g. Harford, 1999). Cash holdings are expected to decrease with an increasing share of inside ownership because inside ownership aligns the interests of shareholder with the interest of management. If holding cash is expensive, managerial ownership will reduce the amount of excess cash held by the company. But as the equity owned by the management reaches a certain limit, the managerial discretion and power increases, which countervails the aforementioned effect. As Opler et al. (1999) state, prior research seems to predict that an inside ownership between 5% and 25-40% causes the fewest agency conflicts.

2.4 Additional Factors Influencing Corporate Cash Holdings

In addition to the factors described in the aforementioned models, there are some more variables which affect corporate cash holding behaviour.

First, country specific factors are important. Tax regulations affect cash holdings be­cause in some countries interest income from liquid assets is taxed at the corporate and at the shareholder level. This makes cash holdings more expensive. Due to the opportunity costs of holding cash, already mentioned in Section 2.1, the interest rate itself affects liquid asset holdings, too. When the liquidity component of the term structure rises, holding cash or cash substitutes induces higher opportunity costs (see Opler et al., 1999). Furthermore, managerial discretion and agency costs are partially country specific factors, too. Different extents of shareholder rights, debtholder rights, or generally the tradition of law and order in a certain country increase managerial discretion and, therefore, corpo­rate cash holdings (see Guney et al., 2003). Moreover, Pinkowitz and Williamson (2001) find strong evidence that bank power is positively related to liquid asset holdings.

To my knowledge, the only study in which cash holdings of New Zealand companies are to some extent compared to cash holdings in other countries, is the investigation by Dittmar et al. (2003). Comparing companies from 45 countries, they find New Zealand companies to have a rather low cash to total assets minus cash ratio. While the median value for all the countries is 6.6%, New Zealand companies have a median of only 1.7%. In their study Dittmar et al. (2003) employ some of the country specific factors mentioned above. Using the shareholder rights index as introduced by La Porta, Lopez-de-Silanes, Shleifer and Vishny (1998), some researchers (see Dittmar et al., 2003; Ferreira and Vilela, 2002; Guney et al., 2003) find evidence that in countries with better shareholder protec­tion cash holdings are less^[3]. This is due to the reduced influence of agency costs and managerial discretion which is consistent with the free cash flow theory. As New Zealand is one of the countries with superior shareholder protection (see La Porta et al., 1998), one can expect the influence of agency costs on corporate cash holdings to be relatively minor in this study. Concerning creditor rights and the tradition of law and order, New Zealand is one of the countries with a higher protection, too. However, the influence of these variables is not quite clear as Guney et al. (2003) and Ferreira and Vilela (2002) find contradictory evidence concerning these two variables.

Second, macroeconomic effects contribute to a company’s cash holding decision. Baum, Caglayan, Ozkan and Talavera (2003) develop a model to examine the influence of ma­croeconomic uncertainty on cash holdings and find that firms tend to become more ho­mogeneous in their cash holding behaviour in times of high macroeconomic uncertainty.

Third, liquid asset holdings vary across different industries. Especially, the financial industry is an outlier regarding cash holdings (see Harford, 1999).

The variables discussed in this section are neither included in the aforementioned models of corporate liquidity nor in the regressions in Section 6.2. Because this paper dis­cusses only the determinants of cash holdings for companies from New Zealand, including country specific variables does not improve the model without having data from another country against which the influence is tested. Due to missing data and a lack of predic- tability, macroeconomic effects and industry classifications are not directly considered, too.

Another variable, which is not reflected in the theoretical models, but is included in the regression analysis of this study, is the inverse of Altman’s Z-score (see Altman, 1968). Following MacKie-Mason (1990), the inverse of the Z-score is used as a proxy for the probability of financial distress. This measure of the likelihood of financial distress is used as well by Kim et al. (1998) in their study on corporate cash holdings. The expected relationship between a firm’s liquid asset holdings and the inverse of Altman’s Z-score is negative because firms with a greater probability of financial distress are expected to hold less cash (see Kim et al., 1998).

2.5 Summary of Theoretical Models

As Table 2.1 shows, the static trade-off model, the financing hierarchy model, and the free cash flow model have partially different predictions on how various variables affect cash holdings (see Ferreira and Vilela, 2002; Opler et al., 1999).

illustration not visible in this excerpt

Table 2.1: Summary of model predictions

However, for several reasons there is no clear prediction which of the models works best and the distinction between the different models is not as clear as Table 2.1 might suggest. The reasons for holding liquid assets by financial managers might be a combination of the different model explained. Especially parts of the free cash flow theory can be part of the other models, too, extending the models in views of agency costs of managerial discretion. While some authors find strong evidence for pecking order behaviour (see e.g. Shyam-Sunder and Myers, 1999), there exists evidence for the importance of the static trade-off model, too (see e.g. Kim et al., 1998; Opler et al., 1999). The free cash flow model is examined in some studies and has as well proven to influence corporate cash holdings (see e.g. Kim et al., 1998).

[...]

---

^[1] For a more detailed discussion of agency costs see Jensen and Meckling (1976).

^[2] According to Jensen (1986), investors would like excess cash to be paid out. Contrarily, Myers and Majluf (1984) argue that cash holdings are valuable for shareholders because they allow firms to take advantage of positive investment opportunities. Pinkowitz and Williamson (2002) examine this contradiction in more detail and find to some extent support for both theories. However, the overall results are in favour of the approach of Myers and Majluf (1984).

^[3] Dittmar et al. (2003) find this evidence after controlling for the country specific differences in market development and dividend payments as described in La Porta, Lopez-de-Silanes, Shleifer and Vishny (1997) and La Porta, Lopez-de-Silanes, Shleifer and Vishny (2000).