The Financial Determinants of Cash Holdin

G Model
IBR-921; No. of Pages 12

International Business Review xxx (2012) xxxxxx
Contents lists available at SciVerse ScienceDirect
International Business Review

journal homepage: www.elsevier.com/locate/ibusrev
The nancial determinants of corporate cash holdings: Evidence from some emerging markets
Basil Al-Najjar *
Middlesex University Business School, Middlesex University, London NW4 4BT, United Kingdom
A R T I C L E I N F O
A B S T R A C T
Article history: Received 29 April 2011 Received in revised form 2 February 2012 Accepted 17 February 2012 Available online xxx Keywords: Brazil Cash holdings China India Russia
This paper investigates corporate cash holdings in developing countries. In particular, we look into the effect of capital structure and dividend policy on cash holdings in Brazil, Russia, India, and China and compare our results with a control sample from the US and the UK. Our sample contains 1992 rms across these countries for the period 20022008. We employ Instrumental Variables analysis to control for the endogeneity of the nancial policies (cash holdings, capital structure, and dividend policy). Our results show some evidence that capital structure and dividend policy affect cash holdings. There are similarities between developed and developing countries on the factors determining corporate cash holdings. The results of our cross-country model provide evidence that capital structure, dividend policy, and rm size are important factors in determining cash holdings. Finally, we show that rms operating in countries with low shareholder protection hold more cash. 2012 Elsevier Ltd. All rights reserved.
1. Introduction Ever since Opler, Pinkowitz, Stulz, and Williamson (1999) investigated the determinants of cash holdings, there has been growing interest in explaining why rms hold cash. In the UK context, Al-Najjar and Belghitar (2011) nd that cash represents, on average, 9% of the total assets. Dittmar and Mahrt-Smith (2007), using US data, nd that around 13% of total assets are cash and near-cash assets. Thus, cash represents a sizeable asset for rms. The determinants of cash holding deserve investigation because cash holding has costs. Firms might hold cash to meet future contingencies but meanwhile, they may not invest in protable projects, with positive NPV. High levels of cash may therefore indicate agency problems between rms management and shareholders (Jensen, 1986). Another important cost of holding cash is the opportunity cost if rms are trading off their protable projects to hold it. According to Scott (1995), institutional factors are cognitive, normative and regulative structures that may affect rms nancial practices such as cash holdings. One of these factors is the socio-economic factor including laws, and actors attitudes which is considered to be weak in many emerging markets relative to that in developed markets such as the US (North, 1990, 2005). This is likely to raise the level of uncertainty in transactions and consequently encourage a range of unproductive practices such as cash retention. Further, slow institutional development (i.e., stock market, bank, and other nancial institutions) may motivate rms to adopt conservative nancial practices (North, 2005). On the other hand, nancial globalization driven by the International Financial Institutions has generated a consensus around the need to raise levels of condence in transactions in the developing world and a tendency to adopt Anglo-Saxon conventions in this area
* Tel.: +44 2084115342. E-mail address: b.al-najjar@mdx.ac.uk. 0969-5931/$ see front matter 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.ibusrev.2012.02.004
Please cite this article in press as: Al-Najjar, B. The nancial determinants of corporate cash holdings: Evidence from some emerging markets. International Business Review (2012), doi:10.1016/j.ibusrev.2012.02.004
G Model

2 B. Al-Najjar / International Business Review xxx (2012) xxxxxx
(Kose et al., 2006). Thus, conicting tendencies exist: national path dependency and in international trend to institutional homogeneity. In this paper, we examine actual rm practice, which may reect the inuence of institutions in national settings but may equally be conditioned by global nancialization and its institutional drivers. We contribute to the literature by investigating the nancial determinants of corporate cash holdings in developing countries, namely Brazil, Russia, India, and China, and compare the results with those based on developed markets: the UK and the US. This allows us to explore corporate cash holdings across countries with different institutional frameworks. We employ both country specic analysis and static panel data estimations. We control for the endogeneity between capital structure, dividend policy and cash holdings. We use a different set of factors that can affect cash holdings for an updated period of time (20022008). Consequently, in this paper we shed more light on developing countries and then we develop a cross-country model to capture the nancial determinants of cash holdings. To the best of our knowledge, this paper is among the rst to concentrate on emerging markets (Brazil, Russia, India, and China). To do so, we use a large sample of 1212 non-nancial rms listed in these countries. The results of our analysis indicate that leverage, dividend payout, liquidity, protability, and rm size impact cash holdings. We examine a cross sectional-time series model across the investigated countries and report that there are country effects on corporate cash holdings decision. The results of the cross-country model show that leverage, dividend payout ratios, and rm size affect cash holdings. The remainder of this paper is organized as follows: Section 2 discuses the theoretical framework; Section 3 develops the hypotheses; Section 4 explains data and methodology; Section 5 highlights the results; Section 6 provides a common model for cash holdings, Section 7 demonstrates further analysis, and nally Section 8 concludes the study. 2. Corporate cash holdings: theory and empirical base Here we provide the theoretical ground of cash holdings. In line with the previous literature we discuss the main theories of cash holdings (see, Opler et al., 1999; Ozkan & Ozkan, 2004). Then, we highlight the ndings of the previous empirical studies. 2.1. Trade-off theory Trade off theory argues that rms maximize their values by considering the marginal costs and marginal benets of holding cash. Under the assumption that managers aim to maximize shareholder wealth, holding cash will bear the cost-ofcarry. This cost is related to the difference between the earnings from holding cash and the interest that rms will pay to fund additional cash (see, Dittmar, Mahrt-Smith, & Servaes, 2003). The benets of holding cash are based on two motives: transaction minimization and precautionary motives. In relation to the former, it is suggested that rms stockpile cash when the rising-costs and the opportunity costs (related to cash decits) are higher (Dittmar et al., 2003; Miller & Orr, 1966; Tobin, 1956). The precautionary motive, based on the effect of asymmetric information on raising funds, suggests that even if rms are able to raise funds from capital markets, they might be reluctant to do so because of market issues (for example if the market is under-pricing the planned securities to be issued). Ozkan and Ozkan (2004) further argue that rms raise cash levels to direct more nancial resources into such investments when the costs of outside nancing are explicitly high. Opler et al. (1999) ascertain the prevalence of an optimal level of cash where the marginal costs of cash shortage match the marginal costs of holding cash. Ferreira and Vilela (2004) argue that holding cash serves to reduce the probability of nancial distress due to unexpected losses. Such rms stockpile cash levels as they are in a better position to direct these resources to investment plans, even if it is hard to obtain funds. Market imperfections are more severe in emerging markets compared to developed markets as well as bankruptcy related costs are signicant in such markets, and hence trade-off theory can explain cash holding decisions in these markets. For example, the ndings of Al-Najjar (2011) and Booth, Aivazian, c Demirgu -Kunt, and Maksimovic (2001) support this argument in emerging market context. There are different proxies, as nancial determinants of cash holdings, used by empirical studies to reect this theory. For example, Al-Najjar and Belghitar (2011), following Ozkan and Ozkan (2004) and Opler et al. (1999) empirically investigate the trade-off theory from cash perspectives by employing leverage, dividend policy, rm size, risk, and asset liquidity. In the same vein, but using EMU data, Ferreira and Vilela (2004) also use liquidity, leverage, growth and size to empirically inspect this theory. We shed more light on the empirical evidence and our nancial factors in Sections 2.3 and 3 below. 2.2. The pecking order theory This theory suggests that there is no optimal level of cash holdings for a rm. Based on asymmetric information, Myers (1984) and Myers and Majluf (1984) suggest that rms follow a pecking order of nancing to minimize costs related to information asymmetry. The order starts with internal sources and rms will use external sources, after the internal sources are exhausted. Myers (1984) proposes that rms favour external funding by debt compared to equity issuance, since debt has lower information costs than equity nancing. Cash can be seen as an outcome of the different nancing and investment decisions proposed by the hierarchal pattern of nancing (Dittmar et al., 2003). Ferreira and Vilela (2004) claim that cash can be used for nancing investments to pay rms debt and in turn stockpile cash. Dittmar et al. (2003) also detect that rms with high level of cash ows are those to distribute dividends, apply for debt nancing, and as a result hoard cash. Based on
G Model

B. Al-Najjar / International Business Review xxx (2012) xxxxxx 3
the previous literature we argue that information asymmetry is also important, and might be more severe in developing countries (see among others, Al-Najjar, 2011; Booth et al., 2001). Different nancial factors, as determinants of cash holdings, have been used by empirical studies to reect this theory. Recently, Al-Najjar and Belghitar (2011) employ leverage and protability as nancial variables that determine the decision to hold cash. In addition, Ferreira and Vilela (2004) use size and cash ow to empirically analyse this theory. We discuss the determinants of cash holdings and the empirical evidence, in more details, in Sections 3 and 2.3 below. To recap, there is no preference in the nance literature, about the best theory that can explain the determinants of cash holdings and thus there is no optimal set of factors that determine the decision to hoard cash. Hence, the aim of our research is to empirically test which variables determine cash holdings for our sample of emerging countries. 2.3. Empirical evidence Numerous studies have focused on cash holdings. These include Opler et al. (1999) using US data; Dittmar et al. (2003) using an international sample of 45 countries; Ozkan and Ozkan (2004) and Al-Najjar and Belghitar (2011) in the UK setting; Dittmar and Mahrt-Smith (2007) in the US; Ferreira and Vilela (2004) within an EMU context; Garcia-Teruel and MartinezSolano (2008) using Spanish SMEs data; Ramirez and Tadesse (2009) using international data. These papers consider developed countries, with two exceptions who use international data (Dittmar et al., 2003; Ramirez & Tadesse, 2009). Dittmar et al. (2003) include Brazil and India but not China or Russia in their sample, and their main interest has been to examine the international corporate governance impact on cash holdings. Ramirez and Tadesse (2009) aim to develop a cross sectional-times series model across 49 countries to investigate the extent to which culture impacts cash holdings. They operationalize culture through the uncertainty avoidance concept which seems likely to capture elements of the organisational/institutional setting as well as fundamental social attitudes. They nd that this uncertainty avoidance variable has a signicant impact on the cross country variations of holding cash. However, Ramirez and Tadesse did not take dividend policy into account in their models. In addition to the determinants of cash holdings, other researchers have investigated the link between cash holdings and rm value. For example, Martinez-Sola, Garcia-Teruel, and Martinez-Sola (2010) use US data and show that there is a concave relationship between cash holdings and rm value. In a similar vein, Tong (2009) nds that cash value is lower in diversied than in non-diversied rms. Pinkowitz, Stulz, and Williamson (2006) detect a weak relationship between cash holdings and rm value in countries that suffer from lower investor protection compared with countries with stronger protection. Foley, Hartzell, Titman, and Twite (2007), using a sample of US multinational rms, examine the importance of holding high cash levels in a taxation context. Their results demonstrate that rms with high levels of repatriation tax increase their cash reserves. Guney, Ozkan, and Ozkan (2003) examine cash holdings in different countries including Japan, France, Germany, and the US. They show that rms with strong shareholder protection are in a better position to hold lower levels of cash. They point out a negative relationship between ownership concentration and cash holdings. Yet, there is limited evidence of corporate cash holdings in relation to developing countries. We aim to bridge this gap in the literature. 3. The determinants of cash holdings: hypotheses development Firm value is not related to rms nancial decisions in the world of no market imperfections (Stiglitz, 1974), which can explain the reason why cash holdings can be irrelevance to rm value (see, Opler et al., 1999). However, these imperfections do exist and are more relevant to emerging markets. Accordingly, we suggest that the theoretical underpinnings determining cash holdings in developed markets will also apply to the emerging market context. In this part of the study, we introduce our hypotheses. We start with leverage, then dividend policy and move afterwards to the other nancial factors included in this study. In addition, we discuss the shareholder protection as a determinant of cash holdings in Section 7. 3.1. Leverage Leveraged rms are more likely to hoard cash due to the higher probability of nancial distress. It is suggested that cash levels decrease with more debt (Baskin, 1987). Accordingly, rms with more liquid assets can covert these assets to cash and in turn hold lower levels of cash (Al-Najjar & Belghitar, 2011; Ozkan & Ozkan, 2004). Opler et al. (1999), Ozkan and Ozkan (2004), and Al-Najjar and Belghitar (2011) argue for a negative link between leverage and cash holdings. We suggest that bankruptcy related costs in emerging markets are also important, since different studies in the emerging market literature nd evidence for these costs (Al-Najjar, 2011; Booth et al., 2001). In a similar vein but supporting the monitoring role of nancial institutions, Ferreira and Vilela (2004) suggest that rms with high level of debt are less able to stockpile cash. This is because they are better monitored if compared to rms with relatively low debt. Hence, based on the previous empirical ndings and the trade off theory, our rst hypothesis is: H1. There is a negative association between leverage and cash holdings.
G Model

3.2. Dividend payments Based on the trade off theory, the association between dividend payments and cash should be negative, since dividendpaying rms can trade off the costs of holding cash by reducing dividend payments. In other words, rms that distribute dividends to their shareholders are more able to raise funds at lower costs when needed by reducing their dividend payments (Al-Najjar & Belghitar, 2011). Similarly, Ozkan and Ozkan (2004:2106) argue that these costs can be avoided for rms facing low internal nancing resources by issuing equity or even reducing payment of dividends. They state that: rms that currently pay dividends can afford to hold less cash as they are more capable of raising funds when needed by cutting dividends. Opler et al. (1999:8) also support this negative relationship: If the rm has a shortage of liquid assets, it can cope with the shortage by either decreasing investment or dividends, or by raising outside funds through security issuances or asset sales. Based on the previous empirical ndings and the trade-off theory, we hypothesize that: H2. There is a negative association between dividends and cash holdings. 3.3. Protability Based on the hierarchal pattern of nancing, cash is an outcome of the nancing and investment activities (Dittmar et al., 2003). Accordingly, protable rms are more able to pay dividends, pay their debt obligations, and stockpile cash. Less protable rms will hold less cash and issue debt to nance their projects. These rms will be reluctant to issue equity because of the high related costs of such issuance (Al-Najjar & Belghitar, 2011; Dittmar et al., 2003; Ferreira & Vilela, 2004). In the same vein, Dittmar et al. (2003:116) argue that rms with low cash ows draw down their cash and issue debt to nance investment, but they refrain from issuing equity because it is too costly. Hence, there is a positive association between rms protability and cash holdings. Based on the empirical ndings and pecking order theory, we hypothesize that: H3. There is a positive association between protability and cash holdings. 3.4. Liquidity It is expected that the costs to convert liquid assets to cash are much lower than other assets. Thus, Firms with more liquid assets can covert these assets to cash and in turn are less likely to hoard cash. The existence of liquid assets will lead rms to be less reliable on capital markets to obtain cash. Ferreira and Vilela (2004), based on trade off theory, argue that in the event of cash shortfall, liquid assets can be easily liquidated and hence they are substitutes for cash. Based on trade-off theory, we argue for a negative relationship between liquidity and cash holdings (Al-Najjar & Belghitar, 2011; Ozkan & Ozkan, 2004) and hypothesize that: H4. There is a negative association between asset liquidity and cash holdings. 3.5. Firm size Small rms are found to hold more cash than their large counterparts because of the high costs of external funds. Large rms are considered to be more diversied than their small counterparts and in turn less prone to bankruptcy related costs (Al-Najjar & Belghitar, 2011) and hence less likely to stockpile cash reserves. Equally it could be argued that large rms have less information asymmetry (if compared to small rms), and so their managers have better exibility in nancial policies and in turn such rms will hold more cash. In a similar vein, Ozkan and Ozkan (2004) argue that if rm size is a proxy for information asymmetry, which reects the external nancing costs, then a negative relationship with cash holdings should be expected. However, if rm size is seen as an index for nancial distress, then small rms are more likely to be liquidated if they suffer from nancial distress. Hence, such rms hold more cash to avoid such distress situation (Ozkan & Ozkan, 2004). Accordingly, we argue that rm size is an important determinant of cash holdings and do not predict the sign of the association between rm size and cash holdings: H5. There is a negative/positive association between rm size and cash holdings. 4. Data and methodology In this section, we start with the importance of our sample; we then offer a description of our data, and nally the econometric model is outlined. 4.1. Sample overview Our sample is emerging markets, namely, Brazil, Russia, India, and China. We select this sample for different reasons. On the macro-economic level, these countries are fast growing economies. They have high foreign direct investment and other multilateral opportunities (HermanMiller, 2010; PriceWaterHouseCoopers, 2010).
G Model

In a recent IMF working paper, Samake and Yang (2011) argue that Brazil, Russia, India, and China (BRIC) have always been considered as key players in the international trade and the global economy. The main reason for such key role is their recorded rapid economic growth in the recent decades. Samake and Yang argue that the role of these countries in world economy is based on three aspects. Firstly, population: the population of the world is around seven billion inhabitants, with three out of seven live in these countries. This shows the signicant contribution of Brazil, Russia, India, and China in the human resource base. Secondly, growth: these countries also have another key contribution to the world economy as their GDP represents the third after the US and the Euro Area. The IMF expects that their GDP will outperform the Euro Area before 2015. Finally, trade: the share of the world exports for these countries have almost tripled over the last two decades, outperforming the US and are getting closer to the Euro Area. From imports point of view, their share has doubled and will soon reach to the US level. Accordingly, these countries have strong integration with the world economy and they are playing a signicant role in the world trade and the nancial ties, especially with low income countries (Samake & Yang, 2011). In the same vein, Wilson and Purushothaman (2003) argue that the role of these countries will be more signicant since they are seen as the new demand and spending power which will compensate the slow expected growth of the more developed countries and thus attracting the multinational rms to invest in them. This is because these countries have the potential to create new multinational rms (Fan, 2008). Enderwick (2009) argues that such emerging markets have a signicant share in the world economy that qualify them to catch up the sales level of the developed economies at lower costs and risks, and thus suppliers are encouraged to invest in them. From the corporate governance point of view, there is a signicant degree of condence in the corporate governance and nancial reporting in these countries. The Financial Standard Foundation indicates that Brazil, India, Russia are enacted and China has intent declared in terms of international governance practice standards. From a nancial auditing point of view, all the countries have intent declared with the International Standards of Auditing (ISA). This shows that nancial reports are clean and investors (banks) can rely on the provided information. Finally, the banking system is seen to follow the international standards for all the countries, with only Brazil as an exception. This gives an indication of banks liquidity and lending procedures in the investigated countries (EstandardsFORUM, 2011). Nevertheless, these countries provide us with very different institutional settings. For example, most listed Chinese rms are state owned; this might have an impact on nancial decisions and the accessibility of external funds. India has liberalized capital markets, with improved shareholder and creditor rights and high levels of foreign inuence in its stock market (Estrin & Prevezer, 2011), and hence better conditions for rms to access the capital market. In Russia, the introduction of the bankruptcy law in 1998 and 2000 provides more transparency and hence lower information opacity. Brazil has good compliance with the international good corporate governance standards that might affect rms nancial decisions (Estrin & Prevezer, 2011). The Bertelsmann Status index for these countries places Brazil and India in the advanced group of countries and Russia (China) in the limited (very limited) group (Bertelsmann, 2010). The extent to which the rule of law is observed in these countries also varies considerably. The Bertelsmann rule of law index indicates that Brazil and India have relatively high scores (7.8 and 7.5, respectively) and that Russia and China are signicantly below these levels (4.3 and 2.3, respectively) (Bertelsmann, 2010). Appendix 1, provides summary statistics the shows the importance of these countries in the world economy. This discussion conrms the importance of our investigated emerging markets in terms of their macro level impact in the world economy and their rich and diversied institutional environments. This also feeds into the signicance of our main aim to investigate if the nancial determinants of cash holdings, which are found important in developed markets, will hold in our sample of emerging markets. 4.2. Data We use DataStream to create the sample based on market capitalization, which consists of 83 rms from Brazil, 93 rms from Russia, 542 rms from India, and 494 in China. These rms are non-nancial. The selection criterion is based on data availability for non-nancial rms in DataStream across the period 20022008. We allow rms to freely enter and exit the market to avoid any survivorship bias in our analysis. Then we compare the results of our main sample with non-nancial rms from the US and the UK, by analyzing a random sample of 576 US rms and 204 rms from the UK. Accordingly, the overall sample is 1992 rms for the period 20022008. The descriptive statistics, reported in Table 1, show that rms in our sample hold low level of cash, as it represents around 5% of the total assets in Russia, 3.5% in China, 3% in India, and 2% in Brazil, compared to 10%, and 8% for the US and the UK, respectively. If we compare our results with previous studies, we nd that average cash holdings in the UK reported by AlNajjar and Belghitar (2011) is 9%, which is similar to our ndings for the UK context. Dittmar et al. (2003) report cash holdings as follows: Brazil is 7%; India is 3%; the UK is 8%, and in the US is 6%. Ramirez and Tadesse (2009) detect that the average cash holdings in Brazil, Russia, India and China are: 9%, 7%, 6%, and 18%, respectively. This difference might be due to the different sample size and the different investigated period. On average, all rms across the countries enjoy good nancial performance with protability more than 80%. On average, leverage is between 20% and 30% across our main sample and 17 20% in the US and the UK. This indicates an almost similar debt pattern (capital structure) in both developing and developed countries in our sample. Similar ndings are reported for liquidity, dividend payout, and rm size. Overall, companies in Russia hold the highest amounts of cash, rely less on debt and pay low dividends. While rms in India, hold relatively high cash, rely on debt and pay high dividends. The case of Chinese rms is different, as they hold cash,
G Model

Table 1 (a) Descriptive statistics and (b) descriptive statistics: whole sample. Country (a) Brazil Russia India China UK US Total Variables (b) Cash Leverage DPO Protability Liquidity Size No. of rms 83 93 542 494 204 576 1992 Cash 0.022 0.048 0.033 0.034 0.101 0.078 Obs 11,692 11,766 10,638 11,780 11,647 11,743 Leverage 0.312 0.218 0.308 0.273 0.231 0.176 DPO 22.571 15.730 17.785 25.532 31.386 12.486 Mean 0.056 0.242 0.195 0.813 1.045 14.137 Protability 0.819 0.857 0.831 0.787 0.810 0.818 Min 0 0 0 0 0.244 0 Liquidity 1.028 1.020 1.035 1.031 1.026 1.075 Size 14.496 17.345 15.155 14.045 13.151 13.494 Max 1 6.900 1 1.439 16.443 21.981
Note: Cash is cash ratio measured by cash to total assets ratio; Lev is leverage ratio, measured by total debt to total assets ratio; DPO is dividend payout ratio measured by dividend per share divided by earning per share; ROE is return on equity ratio measured by net income divided by owners equity; LIQ is liquidity ratio measured by most liquid assets (net of cash) to current liability ratio; Size is the natural logarithm of total assets; Leverage is total debt to total assets ratio; DOP is dividend per share dividend by earning per share. Note: Panel (a) shows the approximate average from the investigated variables in each country.
rely less on debt and pay high dividends. Finally, Brazilian rms hold the lowest level of cash, rely heavily on debt and pay relatively high dividends. Table 2 shows the correlation matrix in which we notice that there are no high correlations among the variables across our sample of countries, in each country and the entire countries, and hence multicolinearity is not a concern, which is conrmed by an average VIF of 1.20 across the models. 4.3. Methodology In order to investigate our hypotheses, we apply different methods, including cross-sectional-time series analysis of the determinants of cash holdings for each country investigated. Then, we test if there is a country effect in determining the decision to hold cash. We subsequently run a time series-cross sectional analysis for the entire sample. The following model is used for the rst stage of our analysis: CASHit b0 b1 LEV it b2 DPOit b3 ROEit b4 LIQ it b5 SIZEit eit where CASH is the cash ratio measured by cash and cash equivalents to total assets ratio; LEV is the leverage ratio, measured by total debt to total assets; DPO is the dividend payout ratio measured by dividends per share divided by earnings per share; ROE is the return on equity ratio measured by net income divided by owners equity; LIQ is the liquidity ratio measured by the most liquid assets to current liabilities; SIZE is the natural logarithm of total assets; e is the error term. For our second stage we investigate the country effect in the cash holdings model, by estimating the following model: CASHit b0 hSDit eit where Dit is a vector of dummy variables representing each investigated country. In so doing, we adopt the approach of Booth et al. (2001) who examine the capital structure decision across developing countries (for more discussion, see Section 5). Finally, we use the following model in the third stage of our analysis: CASHit b0 b1 LEV it b2 DPOit b3 ROEit b4 LIQ it b5 SIZEit h X Dit eit
The variables have the same denitions as above; this model represents the cross sectional time series model to capture the country effects into the model. These models might suffer from endogeneity between the dependent variable and the nancial policies, capital structure and dividend policy.1 As a result, we apply Instrumental Variables (IV) analysis. We use asset tangibility and free cash ows as instruments in our models. Finally, we add year dummies and re-estimate the models
1 We use Hausman test for the endogeneity issue, the results show that there is some evidence of endogeneity in our models, and hence we report the IV analysis to control for it.
G Model

B. Al-Najjar / International Business Review xxx (2012) xxxxxx Table 2 Correlation matrix. Leverage Panel A: Brazil Leverage DPO Protability Liquidity Size Cash Panel B: Russia Leverage DPO Protability Liquidity Size Cash Panel C: India Leverage DPO Protability Liquidity Size Cash Panel D: China Leverage DPO Protability Liquidity Size Cash Panel E: UK Leverage DPO Protability Liquidity Size Cash Panel F: US Leverage DPO Protability Liquidity Size Cash Panel G: whole sample Leverage DPO Protability Liquidity Size Cash 1 DPO 0.081 1 Protability 0.008 0.336 1 Liquidity 0.439 0.105 0.019 1 Size 0.144 0.301 0.479 0.198 1 Cash 0.018 0.117 0.037 0.167 0.01 1 7
0.116 1
0.058 0.177 1
0.492 0.091 0.091 1
0.037 0.192 0.29 0.157 1
0.179 0.013 0.02 0.173 0.16 1
0.193 1
0.065 0.043 1
0.258 0.054 0.01 1
0.081 0.149 0.342 0.271 1
0.111 0.006 0.007 0.264 0.118 1
0.189 1
0.003 0.137 1
0.161 0.041 0.058 1
0.155 0.105 0.202 0.01 1
0.138 0.057 0.003 0.102 0.194 1
0.068 1
0.078 0.098 1
0.406 0.236 0.07 1
0.27 0.097 0.335 0.46 1
0.327 0.122 0.116 0.415 0.371 1
0.127 1
0.04 0.182 1
0.378 0.258 0.213 1
0.24 0.391 0.34 0.519 1
0.28 0.159 0.131 0.389 0.294 1
0.024 1
0.007 0.092 1
0.17 0.04 0.009 1
0.17 0.167 0.319 0.121 1
0.238 0.075 0.053 0.149 0.232 1
Note: Variables are dened in Table 1.
to control for the trends and other un-included confounding effects. However, for parsimony we do not report the time coefcients in the tables. The standard errors are clustered to capture group effects in each country. To sum up, we introduce a static cross sectional-time series IV analysis to investigate the nancial determinants of cash holdings using rms from emerging markets, and then comparing the results with rms from the US and the UK. 5. Regression results Table 3 presents the results of the regression models; there are two models for each country. The rst model is without year dummies and the second model is with them. To recap, the models are cross-sectional time series models to capture
G Model

8 Table 3 IV country specic analysis. Leverage Brazil Brazil (year-dummies) Russia Russia (year dummies) India India (year dummies) China China (year dummies) UK UK (year dummies) US US (year dummies) 0.051 (0.445) 0.057 (0.349) 0.358** (0.020) 0.368** (0.020) 0.363** (0.001) 0.369** (0.002) 0.547** (0.027) 0.400** (0.001) DPO 0.001 (0.128) 0.001* (0.068) 0.006** (0.002) 0.006** (0.002) 0.002 (0.217) 0.002 (0.300) 0.004* (0.099) 0.001 (0.393) Protability 0.108 (0.086) 0.105** (0.049) 0.384** (0.010) 0.404** (0.010) 0.014 (0.713) 0.016 (0.680) 0.577** (0.043) 0.157* (0.102) 0.287 (0.159) 0.330 (0.139)
*
B. Al-Najjar / International Business Review xxx (2012) xxxxxx
Liquidity 0.034 (0.804) 0.060 (0.584) 0.172 (0.566) 0.191 (0.529) 0.371* (0.058) 0.384* (0.065) 0.024 (0.697) 0.012 (0.697) 1.128** (0.044) 1.255** (0.039) 0.249*** (0.000) 0.246*** (0.000)
Size 0.002 (0.447) 0.001 (0.489) 0.003 (0.586) 0.004 (0.517) 0.009** (0.005) 0.009** (0.010) 0.030*** (0.000) 0.009** (0.001) 0.003 (0.997) 0.0047 (0.994) 0.321** (0.009) 0.342** (0.006)
Constant 0.055 (0.781) 0.010 (0.947) 0.007 (0.981) 0.007 (0.981)
Observations 361 361 323 323
0.705** (0.015) 2105 0.712** (0.018) 2105 0.102 (0.480) 0.239** (0.014) 1.474** (0.012) 1.632** (0.011) 3581 3581 803 803 917 917
1.457*** (0.000) 0.005* (0.069) 1.549*** (0.000) 0.005* (0.060)
0.023*** (0.000) 0.516*** (0.000) 0.201*** (0.000) 0.024*** (0.000) 0.517*** (0.000) 0.201*** (0.000)
Note: Variables are dened in Table 1; standard errors are corrected for heteroscedasticity. * Signicant at 10% level. ** Signicant at 5% level. *** Signicant at 1% level.
rm-specic characteristics and the IV analysis is used depending on 2SLS procedure. We discuss the results for each country below. In relation to Brazil, we report inconclusive evidence for the determinants of cash holdings. The results show that there is a negative relationship between protability and cash holdings. This result contradicts the expected positive sign from pecking order theory and H3. We nd a positive relationship between the dividend payout ratio and cash holdings, which contradicts trade-off theory expectations and H2. This might indicate that less protable rms that are able to pay dividends to keep the reputation of paying dividends are not able to obtain extra funds from external sources such as banks and hence hold cash for any contingencies to improve their nancial positions. Leverage, rm size and liquidity are not found to be signicant determinants of cash holdings in the Brazilian case. Regarding Indian rms, the results are consistent with those predicted by nancial theory. Consistent with H1, we report a negative relationship between leverage and cash holdings. Thus, leverage can be viewed as a substitute for holding cash; equally rms with the ability to access external funds are less in need of cash to pay for investments. In line with H4 a negative relationship is found between asset liquidity and cash holdings, and thus rms with more liquid assets can convert these assets to cash and in turn have less need to hold cash (Al-Najjar & Belghitar, 2011; Dittmar et al., 2003). We detect a negative relationship between rm size and cash holdings, and therefore small rms are less able to obtain external funds and are more in need of holding cash. This result is in line with trade-off theory and H5. Finally, we could not nd support for the relationship between cash holdings, on one hand, and dividend payout ratio and protability, on the other. The Chinese perspective shows that leverage, size, and protability are the main determinants of cash holdings. Consistent with theoretical expectations and H1, we detect a negative relationship between leverage and cash holdings: Chinese rms that are more able to obtain debt are less likely to hold cash as they are able to obtain funds externally. Consistent with pecking order theory and H5, we report a positive relationship between rm size and cash holdings and hence large Chinese rms are more diversied and have an increased need to hold cash. Limited evidence is found for the relationship between protability and cash holdings, with a negative sign contradicting the positive expected sign by pecking order theory and H3. Finally, there is no evidence of the impact of dividend payouts and liquidity on cash holdings. The Russian case shows that leverage, dividend payout ratio, and protability determine cash holdings. In line with H1, a negative relationship is found between leverage and cash holdings. In addition, consistent with H2, a negative relationship is found between dividend payouts and cash holdings. Russian rms that are able to access capital markets have less need to hold cash. In line with pecking order theory and H3, a positive relationship is reported between protability and cash holdings. This indicates that protable rms are more able to pay dividends and accumulate cash holdings (Al-Najjar & Belghitar, 2011). Finally, no evidence is reported for the impact of rm size and liquidity on cash holdings in Russia. The results of our controlled samples from the UK and the US show almost similar results to those reported for our sample of emerging countries, especially in the case of the UK. We nd that leverage, dividend, and liquidity are the determinants of cash holdings with the expected signs. However, in the case of our US sample we report contradictory signs for dividends, protability, and liquidity. After we control for year xed effects, we conrm the previous results across the investigated countries. In particular, leverage, dividend payout ratio, protability, rm size, and liquidity are important drivers of cash holdings in our main sample.
G Model

Overall, the nancial factors that affect cash holdings in developing countries are similar to those that affect cash holdings in developed countries: leverage, dividend payout, protability, asset liquidity, and rm size. 6. A common model for cash holdings across developed and developing countries Our results reported in Table 3 show, to some extent, that our sample of emerging countries shares the same determinants of cash holdings as suggested by the developed markets. However, some variables change in signs or signicance across the countries. There are different reasons behind these differences: for example, the number of rms varies across the countries, as well as different institutional settings. Similar ndings are reported by Booth et al. (2001). We follow the approach of Booth et al. and run the regression of cash holdings against the dummy variables that represent the countries. We exclude the dummy for the UK and keep the US in our sample to maximize the number of observations. This model represents the hypothesis that if we know the nationality of the rms, we can know the determinants of cash holdings. Then, the model will be compared with the cross sectional-time series model for all the countries. Table 4 shows the results for the country dummies, we report two models one with year dummies and the other without them. It can be seen from Table 4 that all the dummies are highly signicant. Around 8% of the variation of the dependent variable can be explained by knowing the rms nationality. This nding indicates that cash holding decisions are related to different industrial and rm-specic factors (such as dividend decisions). Therefore, the inclusion of the rm-specic variables will result in lower effect of country-dummies. Equally, it can be argued that the results of these differences are related to structural or institutional aspects, including the ability to access nancial institutions funds or to access the capital market. In this case the nancial factors should vary and remain signicant. Finally, these differences can be related to missing information or wrong sampling (see for example, Booth et al., 2001). In our nal stage, we estimate two models, one with dummy-variables and the other with year and country dummies. We use the IV model with the 2SLS approach to estimate our cross sectional-times series models. The results are reported in Table 5. The results of the cross sectional-times series model show that leverage is negatively related to cash holdings. This result is consistent with the reported ndings in our country specic models, suggesting that rms with more ability to obtain debt
Table 4 Regression with country dummies. Model 1 Brazil Russia India China US Constant Year dummies Observations R2 0.089*** (0.000) 0.063*** (0.000) 0.077*** (0.000) 0.076*** (0.000) 0.032*** (0.000) 0.110*** (0.000) No 23041 0.083 Model 2 0.089*** (0.000) 0.062*** (0.000) 0.076*** (0.000) 0.076*** (0.000) 0.0322*** (0.000) 0.105*** (0.000) Yes 23041 0.085
Note: Variables are dened in Table 1; standard errors are corrected for heteroscedasticity. *** Signicant at 1% level.
Table 5 Cross section time series regressions. Model 1 Leverage DPO Protability Liquidity Size Brazil Russia India China US Constant Year dummies Observations 0.595** (0.003) 0.006*** (0.000) 0.066 (0.190) 0.126 (0.594) 0.010* (0.061) 0.023 (0.511) 0.133*** (0.000) 0.0687** (0.008) 0.027 (0.397) 0.108*** (0.000) 0.318 (0.247) No 17383 Model 2 0.592** (0.002) 0.0061*** (0.000) 0.072 (0.130) 0.119 (0.611) 0.0098* (0.060) 0.0229 (0.512) 0.130*** (0.000) 0.065** (0.010) 0.021 (0.497) 0.1072*** (0.000) 0.298 (0.270) Yes 17383
Note: Variables are dened in Table 1, Models 1 and 2 are cross sectional-times series regressions; Models 3 and 4 are regressed using 2SLS. Standard errors are corrected for heteroscedasticity. * Signicant at 10% level. ** Signicant at 5% level. *** Signicant at 1% level.
G Model

nancing, are less in need of holding cash. This result is in line with Baskin (1987), Kim, Mauer, and Sherman (1998), and AlNajjar and Belghitar (2011). In addition, a negative relationship is found between dividend payout ratio and cash holdings. Thus, the greater the ability to access capital markets, the less the need to hold cash. This result is consistent with Ozkan and Ozkan (2004), Ferreira and Vilela (2004), and Al-Najjar and Belghitar (2011). Moreover, we detect a positive relationship between rm size and cash holdings and hence large rms are more able to diversify their opportunities and are more in need to hoard cash. Finally, we could not nd any support for the impact of protability and asset liquidity on cash holdings. These results are not signicantly different from those reported in Table 3. The country dummy-variables have dropped in their signicance as only three out of 5 dummies are statistically signicant. The models exclude the dummy variable for the UK. Accordingly, we suggest that the results of the cross sectional-times series model support the argument that the differences in decisions on holding cash are related to institutional and rm-specic factors. When we control for year dummies, we report similar results to those reported in Model 1, and thus we argue that rm specic factors are important in determining cash holdings. More specically, leverage, dividend payout ratios and rm size affect the cash holding decision. 7. Further analysis Our ndings show a cross-country effect. To shed more light on this issue, we investigate in more details the institutional settings of our main sample. To do so, the study includes two variables to measure the impact of shareholder right and the legal system in emerging market context. Here, the theoretical base is the agency cost view of cash holdings. The agency relationship is a result of the separation between management and ownership of the rm. There are some advantages of this separation, including the ability of ownership to change without impacting operations, and the possibility of hiring experts to act as managers (Jensen & Meckling, 1976). However, managers might have their own objectives distinct from those of shareholders, and hence agency problems exist between the agents (managers) and principals (shareholders). Jensen and Meckling (1976) dene the agency relationship as a contract whereby owners delegate managers to act on their behalf. Jensen (1986) argues that rms accumulate free cash ows since managers can control this cash. Managers can hold cash to minimize the likelihood of nancial distress, and use cash to invest in projects that might not be in the interest of shareholders (see for example, Shleifer & Vishny, 1997). From the agency costs perspective, managers that are less concerned to meet shareholders expectations are those who are willing to hold more cash and invest in unprotable projects (negative NPV). In order to investigate agency theory, we adopt Dittmar et al.s (2003) approach, in which they argue that shareholders in a good legal protection environment can force management to reduce cash levels. Hence, we investigate how legal protection will affect cash holdings across countries. Another possible explanation for this relationship is that rms operating in low shareholder protection hoard cash since capital markets are not highly developed and rms nd such markets difcult to approach. Hence, we hypothesize that: H6. There is a negative relationship between shareholder protection and cash holdings. To test H6, we run the following model: CASHit b0 b1 SHARERIGHT b2 DPOit b3 LEV it b4 ROEit b5 LIQ it b6 SIZEit eit where, SHARERIGHT is a dummy variable that takes one if the country is considered to have high shareholder right, and zero otherwise. We replace the variable with COMMON which is a dummy variable to reect the common law in our main sample.
Table 6 The effect of shareholder right. Model 1 Shareright Common Leverage DPO Protability Liquidity Size Constant R2 Observations 0.004
***
Model 2 (0.007) 0.004 (0.016) 0.033*** (0.003) 0.0004*** (0.248) 0.026 (0.150) 0.027 (0.289) 0.0098**** (0.0000) 0.006 (0.570) 0.03 6480
**
Model 3 0.0019 (0.317) 0.033*** (0.000) 0.0004 (0.248) 0.029 (0.218) 0.0274 (0.249) 0.030*** (0.000) 0.007 (0.570) 0.03 5913
0.0361*** (0.000) 0.001 6480
Note: Variables are dened in Table 1, Models 1 and 2 use shareholder right dummy to capture the high share holder right countries; Model 4 uses common law dummy to capture the common law country are regressed, the simple regression model for the common law model is not reported because it was not signicant, the coefcient of interest is 0.0011. Standard errors are corrected for heteroscedasticity. ** Signicant at 5% level. *** Signicant at 1% level.
G Model

Shareholder right dummy, takes the value of one if rms are listed in Brazil and India (Dittmar et al., 2003), and zero otherwise. The common law dummy is for rms listed in India, since it is the only common law country in our main sample. We expect the coefcients of these variables to be negative and signicant. Table 6 shows three models. The coefcients of our main variable in the rst two models are negative and statistically signicant. This is consistent with our expectation and the ndings of Dittmar et al. (2003). When we replace the shareholder right variable by the common law variable, the sign of the coefcient is consistent with our expectations but is not statistically signicant. Hence, we support H6 and conrm that, in the emerging market context, rms operating in countries with less shareholder protection hold more cash. Another notable result is that the other nancial variables have dropped in their signicance, indicating that cash holdings can be seen as an outcome of weak shareholder protection compared to the other investigated nancial determinants. 8. Summary and conclusions In this study, we investigate cash holdings in emerging markets that are different in governance and institutional framework from the US and the UK. When we combine both country and cross-country analyses about cash holdings, we gain more understanding about the debate of why rms hold cash. Furthermore, investigating this debate in emerging markets emphasizes the importance of the strategic decision of holding cash, which has been under-researched or partially explored in the previous literature. Accordingly, our main aim in this study is to provide new evidence on the nancial determinants of cash holdings in emerging markets. The sample includes non-nancial rms from Brazil, Russia, India, and China, which represent the biggest emerging markets as they are among the top 20 countries in terms of the GDP. Our results provide further insights into the decision of holding cash from international context. We detect that for both developed and emerging market rms, leverage, dividend payout, liquidity, protability, and rm size impact cash holdings. It is worth noting that these results might be subject of scepticisms, since some of the independent variables have changed in signicance across the countries, and some have an overall low impact. This is expected because these countries have different institutional settings. Booth et al. (2001) reach to a similar result when they investigate capital structure in developing countries. Booth et al. (2001) and Aivazian and Booth (2003) use India in their sample. Dittmar et al. (2003) examine Brazil, as well as India, and Ramirez and Tadesse (2009) use Brazil, Russia, India, and China in their sample when they investigate cash holdings. Their results show some evidence that nancial theory is applicable in international contexts. Hence, our results are comparable to those reported in the previous literature and provide further insights about the nancial determinants of cash holdings across countries. We adopt a uniform model for developed and developing countries to capture the impact of rm-driven factors. The cross sectional-time series model shows that the investigated nancial factors affect cash holdings. In particular, leverage, dividends and rm size are found to be important nancial determinants of cash holdings. The industrial and institutional settings are the main reasons behind the differences in cash holding decisions across our sample. In order to further investigate this issue, we run additional analysis and provide supporting evidence that, from emerging markets context, rms operating in countries with low shareholder protection hold more cash. One reason is that such countries have weak capital markets for rms to approach for funding. Overall, we provide further evidence that trade-off theory, pecking order theory and agency cost theory play important roles in understanding nancial decisions such as cash holdings in developing countries. Our results show that the factors determining cash holdings in both emerging markets and more developed countries are largely similar. In particular, leverage, dividend payout, protability, asset liquidity, and rm size have an impact on cash holdings. In general, from international perspective, this study has an important implication since it shows that even if emerging markets differ in nancial and governance structures yet they share the same nancial determinants. Hence, rms in such countries follow almost similar patterns in managing their cash holdings. There is still room for further discussion about corporate cash holdings in developing countries by investigating the internal corporate governance factors that might impact nancial decisions. Board characteristics, audit features, and CEO characteristics all require investigation. Appendix A Descriptive statistic of Brazil, Russia, India, and China for 2010.
Country Brazil Russia India China GDP (US$) 2,087,890 1,479,819 1,729,010 5,878,629 Exports %of GDP 10% 29% 18% 29% Imports % of GDP 11% 20% 25% 25% Ination 7.5% 11.4% 9.6% 5.8% Corruption Index 3.7 2.1 3.3 3.5 Governance Enacted Enacted Enacted Intent declared International Standards of Audit Intent Intent Intent Intent declared declared declared declared Civil/Common Law Civil Civil Common Civil
Note: The information about GDP, exports, and imports are from the World Bank statistics; Governance is the extent to which the country follows the international corporate governance standards and International standard of audit reects the extent to which the country follows these standards provided from EstandardsFORUM database.
G Model

References
Al-Najjar, B. (2011). The interrelationship between capital structure and dividend policy: Empirical evidence from Jordanian data. International Review of Applied Economics, 25(2), 209224. Al-Najjar, B., & Belghitar, Y. (2011). Corporate cash holdings and dividend payments: Evidence from simultaneous analysis. Managerial and Decision Economics, 32(4), 231241. Aivazian, V., & Booth, L. (2003). Do emerging market rms follow different dividend policies from US rms? Journal of Financial Research, 26(3), 371387. Baskin, J. (1987). Corporate liquidity in games of monopoly power. Review of Economics and Statistics, 69, 312319. Bertelsmann. (2010). The country status index [online]. Retrieved from: http://www.bertelsmann-transformation-index.de/en/bti/ranking/. Accessed 29.11.11. c Booth, L., Aivazian, V., Demirgu -Kunt, A., & Maksimovic, V. (2001). Capital structures in developing countries. Journal of Finance, 56, 87130. Dittmar, A., & Mahrt-Smith, J. (2007). Corporate governance and the value of cash holdings. Journal of Financial Economics, 83, 599634. Dittmar, A., Mahrt-Smith, J., & Servaes, H. (2003). International corporate governance and corporate cash holdings. Journal of Financial and Quantitative Analysis, 38(1), 111133. Enderwick, P. (2009). Responding to global crisis: The contribution of emerging markets to strategic adaptation. International Journal of Emerging Markets, 4(4), 358374. EstandardsFORUM. (2011). Country proles [online]. Retrieved from: <http://www.estandardsforum.org/>. Accessed 20.03.10. Estrin, S., & Prevezer, M. (2011). The role of informal institutions in corporate governance across BRIC countries. Asia Pacic Journal of Management, 4167. Ferreira, A., & Vilela, S. (2004). Why do rms hold cash? Evidence from EMU Countries. European Financial Management, 10(2), 295319. Fan, Y. (2008). The rise of emerging market multinationals and its impact on marketing. Marketing Intelligence & Planning, 26(4), 353358. Foley, C. F., Hartzell, J. C., Titman, S., & Twite, G. (2007). Why do rms hold cash? A tax-based explanation. Journal of Financial Economics, 86, 579607. Garcia-Teruel, P., & Martinez-Solano, P. (2008). On the determinants of SME cash holdings: Evidence from Spain. Journal of Business Finance & Accounting, 35, 127 149. Guney, Y., Ozkan, A., & Ozkan, N. (2003). Additional international evidence of corporate cash holdings. Working paper. University of York. Jensen, M. C. (1986). Agency costs of free cash ow, corporate nance, and takeovers. American Economic Review, 76, 323329. Jensen, M., & Meckling, W. (1976). Theory of the rm: Managerial behaviour, agency costs and ownership structure. Journal of Financial Economics, 3, 305360. HermanMiller. (2010). Culture and work styles in the BRIC countries research summary. Retrieved from: <http://www.hermanmiller.com/MarketFacingTech/hmc/ research/research_summars/assets/wp_BRIC_Countries.pdf>. Accessed 15.04.11. Kim, C., Mauer, D., & Sherman, A. (1998). The determinants of corporate liquidity: Theory and evidence. Journal of Financial and Quantitative Analysis, 33, 335359. Kose, M. A., Prasad, E., Rogoff, K., & Wei, S. -J. (2006). Financial globalization: A reappraisal. IMF working paper 06/189. Martinez-Sola, C., Garcia-Teruel, P., & Martinez-Sola, P. (2010). Corporate cash holdings and rm value. Multinational Finance Conference. Miller, M. H., & Orr, D. (1966). A model of the demand for money by rms. Quarterly Journal of Economics, 80(3), 413435. Myers, S. C. (1984). The capital structure puzzle. Journal of Finance, 39(3), 575592. Myers, S. C., & Majluf, N. S. (1984). Corporate nancing and investment decisions when rms have information that investors do not have. Journal of Financial Economics, 13, 187221. North, D. (1990). Institutions institutional change and economic performance. Cambridge University Press. North, D. (2005). Understanding the process of economic change. Princeton University Press. Opler, T., Pinkowitz, L., Stulz, R., & Williamson, R. (1999). The determinants and implications of corporate cash holdings. Journal of Financial Economics, 52, 346. Ozkan, A., & Ozkan, N. (2004). Corporate cash holdings: An empirical investigation of UK companies. Journal of Banking and Finance, 28, 21032134. Ramirez, A., & Tadesse, S. (2009). Corporate cash holdings, uncertainty avoidance, and multinationality of rms. International Business Review, 18, 387403. Pinkowitz, L., Stulz, R. M., & Williamson, R. (2006). Does the contribution s of corporate cash holdings dividend to rm value depend on governance? A crosscountry analysis. Journal of Finance, 61(6), 27252751. PriceWaterHouseCoopers. (2010). Economic views [online]. Retrieved from: <http://www.pwc.com/dk/da/publikationer/assets/economic-view-bric.pdf>. Accessed 08.04.11. Samake, I., & Yang, Y. (2011). Low-income countries BRIC Linkage: Are there growth spillovers? IMF working paper, WP/11/267. Retrieved from: <http:// www.imf.org/external/pubs/ft/wp/2011/wp11267.pdf>. Accessed 29.11.11. Scott, R. (1995). Institutions and organisations. London: Sage. Shleifer, A., & Vishny, R. W. (1997). A Survey of corporate governance. Journal of Finance, 52(2), 737783. Stiglitz, J. E. (1974). On the irrelevance of corporate nancial Policy. The American Economic Review, 64(6), 851866. Tobin, J. (1956). The interest elasticity of transactions demand for cash. Review of Economics and Statistics, 38(3), 241247. Tong, Z. (2009). Firm diversication and the value of corporate cash holdings. Journal of Corporate Finance, 142. Wilson, D., & Purushothaman, R. (2003). Global economics paper no. 99: Dreaming with BRICs: The path to 2050. Retrieved from <http://www2.goldmansachs.com/ ideas/brics/bricsdream.html>. Accessed 29.11.11.

The Financial Determinants of Cash Holdin

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Financial Determinants of Cash Holdin

Uploaded by

Copyright:

Available Formats

G Model

IBR-921; No. of Pages 12

Contents lists available at SciVerse ScienceDirect

International Business Review

IBR-921; No. of Pages 12

IBR-921; No. of Pages 12

IBR-921; No. of Pages 12

IBR-921; No. of Pages 12

IBR-921; No. of Pages 12

IBR-921; No. of Pages 12

0.492 0.091 0.091 1

0.037 0.192 0.29 0.157 1

0.179 0.013 0.02 0.173 0.16 1

0.258 0.054 0.01 1

0.081 0.149 0.342 0.271 1

0.111 0.006 0.007 0.264 0.118 1

0.161 0.041 0.058 1

0.155 0.105 0.202 0.01 1

0.138 0.057 0.003 0.102 0.194 1

0.406 0.236 0.07 1

0.27 0.097 0.335 0.46 1

0.327 0.122 0.116 0.415 0.371 1

0.378 0.258 0.213 1

0.24 0.391 0.34 0.519 1

0.28 0.159 0.131 0.389 0.294 1

0.17 0.04 0.009 1

0.17 0.167 0.319 0.121 1

0.238 0.075 0.053 0.149 0.232 1

Note: Variables are dened in Table 1.

IBR-921; No. of Pages 12

B. Al-Najjar / International Business Review xxx (2012) xxxxxx

Constant 0.055 (0.781) 0.010 (0.947) 0.007 (0.981) 0.007 (0.981)

Observations 361 361 323 323

1.457*** (0.000) 0.005* (0.069) 1.549*** (0.000) 0.005* (0.060)

IBR-921; No. of Pages 12

IBR-921; No. of Pages 12

0.0361*** (0.000) 0.001 6480

IBR-921; No. of Pages 12

IBR-921; No. of Pages 12

You might also like