Internationalization and working capital management efficiency: the role of
country economic development and price stability
Abstract: I study effects of internationalization on working capital management (WCM)
efficiency. My results show that internationalization is generally harmful for multinational firms (MNCs). WCM efficiency of global firms with large share of foreign sales does not differ from local firms. However, when I correct my model for the country economic development, I find that MNCs from less economically developed countries can benefit from internationalization. I also find a severe impact of price instability on firms; however, this does not differ between MNCs and local firms. I show that MNCs under certain circumstances can benefit from international short-term funding and investing opportunities, trade credit provisions provided by international business partners, multilateral netting systems and centralized cash management. It mostly depends on the economic development of a domestic country.
Key words: working capital management efficiency, cash conversion cycle, firm
internationalization, country economic development, price stability
Name: Martin Durica Study programme: MSc IFM
2
1. Introduction
History of capital management research shows the strong focus on the long-term capital management; however, the key to the profitability of firms may be hidden in the short-term working capital management (Boschker, 2011). Chang and Noorbakhsh (2009) indicate that according to the World Scope data for the period 1995 to 2004, 15% of total assets of multinational firms (MNCs) were in form of cash, cash equivalents or highly liquid assets. Smith (1987) demonstrates the importance of working capital management (WCM) by claiming that it is crucial to enhance the firm market value. Moyer, McGuigan, and Rao (2017) in addition emphasize its essential role in the financing of firms, which is a consequence of requirements imposed on firms by suppliers of finance.
In developing my arguments, I review the current literature on WCM efficiency. Researchers who study WCM efficiency mostly examine how it impacts firm’s profitability and firm value within one country. Indeed, Baños-Caballero, García-Teruel, and Martínez-Solano (2010) study the relationship between WCM and corporate performance on a sample of firms from the United Kingdom. Similar analysis on other countries can be found in numerous research papers (see, e.g., Aktas, Croci, and Petmezas, 2015; Chang, 2018; Deloof, 2003; Schiff and Lieber, 1974).
3
enhances firm performance. Apart from the managerial implications, this also affects academic research; including my research paper.
Nevertheless, WCM efficiency topic can be approached from various angles. Dorsman and Gounopoulos (2008), in discussion on working capital in MNCs, claim that there are differences in WCM even among countries. Preve and Sarria-Allende (2010) suggest that certain macroeconomic factors can explain these country-specific differences. Westerman (2015), using the research of Preve and Sarria-Allende (2010), mentions monetary constraints, policy rates of central banks (CBs), inflation and financial crisis as the warning indicators on headquarters level.
Different from the previous research, but considering earlier suggestions of Preve and Sarria-Allende (2010) and Westerman (2015); the main focus of my analysis is to explain how internationalization impacts working capital management efficiency; how country economic development moderates the relationship between internationalization and working capital management efficiency; and how price stability moderates the relationship between internationalization and working capital management efficiency. Thus, the main question of this research is: Does firm internationalization matter for working capital management efficiency?; and the sub-questions are: Does country economic development moderate the relationship between
firm internationalization and working capital management efficiency? and Does price stability moderate the relationship between firm internationalization and working capital management efficiency?.
4
missing or incorrect values are also omitted. OLS unbalanced panel regressions are performed to analyze the data.
My hypotheses draw a complete picture of the relationship between internationalization and WCM efficiency. The first set of tests partially supports my hypothesis that MNCs have less efficient WCM than local firms. The only exception are the most global MNCs with more than 50% foreign sales, which do not significantly differ from local firms. These results are in line with Bekaert and Hodrick (2017), who suggest that MNCs do not benefit from internationalization because they must face additional obstacles such as high capital transfer costs, various tax requirements or exchange rate fluctuations.
The second set of results reveals the moderating role of country economic development. My empirical findings, consistent with my hypothesis, show that MNCs headquartered in emerging countries benefit from their international status and improve WCM efficiency thanks to their international opportunities. However, their advantage fades in countries with a higher GDP per capita. This is all in line with theoretical claims of Bekaert and Hodrick (2017). Sensitivity analysis performed on the sample of less developed countries gives further support to my findings.
The third and last set of results shows that MNCs from high inflation countries cannot benefit from their international operations and significantly improve their WCM efficiency. These findings are conflicting with my hypothesis. Results show that, consistent with De Grauwe (2018), price instability has a negative effect on the WCM efficiency of MNCs as well as local firms. Sensitivity analysis performed on the sample of high inflation countries adds further support for my findings.
5
discusses the results; and Section 5 concludes the research paper, explains managerial implications, denotes limitations and gives suggestions on further research.
2. Background and hypotheses
2.1 Working capital management efficiency
WCM is imperative for firms. Investments in accounts receivable and inventories may represent substantial share of assets (Baños-Caballero, García-Teruel, and Martínez-Solano, 2010); and WCM may also play a crucial role in firm’s financing (Moyer, McGuigan, and Rao, 2017). Cuñat (2008) finds that more than 40% of total debt of firms from the United Kingdom is represented by trade credit.
Working capital, as defined by Bekaert and Hodrick (2017), consists of cash, inventory and other net short-term assets. Westerman (2015) proposes that WCM is related to accounts receivable, accounts payable and inventory. WCM has several roles and goals. According to Platt (2010), the main role of WCM is to monitor working capital and to secure enough resources to pay the bills and short-term debt. Jensen (1986) also argues that some managers may hold high cash reserves to achieve an increase in firm’s assets; and consequently, increase the power of firms over their investments (Ferreira and Vilela, 2004).
6
Aktas, Croci, and Petmezas (2015), in their research of US firms for the period 1982 to 2011, find that an optimal level of working capital exists. Firms that manage to reach an optimal level subsequently improve their financial performance. Changes in investments in working capital serve as a channel to reach an optimal level. In other words, firms that manage to use their corporate resources in the right way benefit the most. However, Howorth and Westhead (2003) find that these resources are limited. Thus, an optimal level of working capital may not always be achievable.
2.2 Working capital management efficiency and its traditional determinants
Firm’s size, sales growth, profitability, debt and industry in which a firm operates are factors often discussed in theoretical literature as determinants of WCM efficiency (see, e.g., Baños-Caballero, García-Teruel, and Martínez-Solano, 2010; Westerman, Wu, Wiekens, Noordmans, and Laseur, 2016). Baños-Caballero, García-Teruel, and Martínez-Solano (2010) find that a firm’s size has a positive effect on its WCM efficiency. The same research provides evidence on the opposite effect of a firm’s profitability and sales growth. Thus, these two factors have a negative impact on WCM efficiency. Wasiuzzaman and Arumugam (2013), using a sample of 192 Malaysian firms, find that firms invest more in working capital when they have a low leverage.
2.3 Internationalization and working capital management efficiency
7
Moreover, Bekaert and Hodrick (2017) propose that multilateral netting systems serve as a tool for MNCs to save on transaction costs; centralized cash management diminishes precautionary demands for cash; precise timing of payments enhances liquidity; and MNCs have an option to move funds from affiliates with high opportunity costs of net working capital to affiliates with lower opportunity costs.
Therefore, considering the previous arguments, I expect to find a positive relationship between WCM efficiency and internationalization:
Hypothesis 1a: Internationalization has a positive effect on working capital management efficiency.
Bekaert and Hodrick (2017) also summarize impediments which MNCs need to overcome on their way to enhanced WCM efficiency. MNCs need to deal with additional costs related to capital transfers from affiliates to parents, foreign exchange (Forex) market operations and taxes paid to foreign tax authorities. These issues require efficient dividend policies, transfer pricing strategies, tax planning, and timing of payments to take advantage of Forex rates changes. Inefficiency may lead to poor performance. Bekaert and Hodrick (2017) also claim that MNCs should spread their investments and debt across several banks. They propose that even the largest banks may default. Diversification across banks consequently increases expenditures and demand for cash.
8
described by Kubick, Lynch, Mayberry, and Omer (2016), aggressive tax and transfer pricing strategies leading to tax evasions may attract an investigation by local authorities.
Less common, but even more intriguing question is whether a longer CCC can have positive effects on firms. Even though I do not study the effect of WCM efficiency on profitability, it is important to see the results of these studies. While Aktas, Croci, and Petmezas (2015) find a concave relationship and most of the papers find a positive effect of WCM efficiency on profitability (see, e.g., Shin and Soenen, 1998); empirical findings of Gill, Biger, and Mathur (2010) suggest that higher investments in WCM, i.e. a longer CCC, can occasionally lead to higher profitability.
Presented arguments suggest some likelihood of a negative relationship between internationalization and working capital management efficiency:
Hypothesis 1b: Internationalization has a negative effect on working capital management efficiency.
2.4 The role of country economic development
9
Capturing the nature of emerging markets, other issues may arise and directly affect financial management. Agyei-Mensah (2010) studies these issues on a sample of Ghanaian firms. He argues that poor education, missing managerial skills or no access to accounting software negatively affect WCM efficiency. He claims that owners and top managers with lack of financial skills are often themselves an obstacle to efficient WCM. Tewolde and von Eije (2007), who study trade credit on a sample of Eritrean firms, partially support this view as they also assign responsibility to managers; however, they claim that the shortcomings are rather caused by the selfishness of managers than by the country economic development. Moreover, Mongrut, O’Shee, Zavaleta, and Zavaleta (2012) find that even among developing countries from the same region, differences in WCM may be significant. Among other factors, they mention country-specific risk as one of the determinants of efficient WCM.
Based on the provided evidence and following mainly the reasoning of Bekaert and Hodrick (2017) and Desai, Foley, and Forbes (2008), I expect country economic development to have a negative impact on the relationship between internationalization and WCM efficiency. I expect that MNCs from developing countries can benefit from their international status even more despite the issues they have to face in their country of origin. The second hypothesis stands as follows:
Hypothesis 2: Country economic development has a negative effect on the relationship between internationalization and working capital management efficiency.
2.5 The role of price stability
10
Depending on the mandate of a CB, price stability is either a solitary ultimate target or one of the ultimate targets of a monetary policy. Countries which do not belong to a monetary union can use various instruments to affect the price stability. However, countries which belong to a monetary union drop these instruments.
De Grauwe (2018) emphasizes that price instability, i.e. high inflation, has direct negative effects not only on purchasing power, budget deficits, country debt or long-term interest rates, but also on economic growth and employment. All of this has an impact on firms. Moreover, debt holders benefit and creditors suffer as a consequence of growing inflation. Thus, the impact on firm’s WCM efficiency depends on the aggregate value of its short-term debt, receivables and payables on its balance sheet. However, De Grauwe (2018) also claims that deflation would be even worse as the value of debt, including short-term debt, would increase.
Based on the arguments of De Grauwe (2018), I assume that high inflation can lead to various problems for firms. However, due to the access to foreign financial markets, MNCs have an enhanced ability to overcome financial constraints (Desai, Foley, and Forbes, 2008). As previously discussed, MNCs can also take an advantage of centralized cash management or a better ability to invest short-term surpluses and borrow short-term deficits (Bekaert and Hodrick, 2017). Madura and Whyte (1990) also claim that MNCs can benefit from their internationally diversified operations and reduce country-level risk.
11
Hypothesis 3: Price stability has a negative effect on the relationship between internationalization and working capital management efficiency.
3. Methodology
3.1 Data collection
I perform Ordinary Least Square (OLS) unbalanced panel regressions in this research paper on a total sample of 16,474 observations. All firm-related data are obtained from Datastream (Thomson Reuters). Data on GDP per capita (GDP) and inflation (Inflation) come from The World Bank databases. Two-digit ISO country codes (ISO) enable me to match data from the two sources. Perfect comparability of events is secured by all values being stated in US dollars. The sample covers an extensive 10-year period 2006 to 2015. This period sufficiently covers years of economic prosperity and years of economic crisis. The data also include inactive firms to avert survivorship bias.
Considering the limits of obtained data, in case of missing information or incorrect entries, I omit the events from the sample. Due to the different characteristics of financial firms, SIC codes (SIC) from 6000 to 6999 are also left out. Financial firms face different regulations; differ in capital structure; and their complexity and opaqueness lead to greater information asymmetry (De Haan and Vlahu, 2016). Stated differences consequently decrease firm value and may increase agency problems. Firm level data are winsorized on a 5% level to reduce the effect of outliers.
12
researchers and can be seen in many corporate finance related research papers (see, e.g., Bruno and Shin, 2014; Aktas, Croci, and Petmezas, 2015).
3.2 Variables
3.2.1 WCM efficiency as a dependent variable
Past research papers use various ways how to measure WCM efficiency. Baños-Caballero, García-Teruel, and Martínez-Solano (2010) employ the Net Trade Cycle as a primary measure of WCM efficiency. Additionally, they employ the square of Net Trade Cycle as a robustness test. However, probably the most common measure of WCM efficiency is the Cash Conversion Cycle (see, e.g., Martínez-Solano and García-Teruel, 2007; Westerman, Wu, Wiekens, Noordmans, and Laseur, 2016).
In my research paper, as the dependent variable, i.e. WCM efficiency, I use two alternatives. As a primary proxy, I use the Cash Conversion Cycle (CCC) measured in days similar as Westerman, Wu, Wiekens, Noordmans, and Laseur (2016) and Martínez-Solano and García-Teruel (2007). The
CCC comprises Inventory Conversion Period, Receivables Conversion Period and Payables
Conversion Period. Following Baños-Caballero, García-Teruel, and Martínez-Solano (2010), I also use the Net Trade Cycle (NTC) measured in days as an alternative proxy. The NTC differs from the CCC by replacing the Cost of Goods Sold in a denominator by Total Sales in all its three components. Thus, the NTC tends to be slightly different from the CCC as a day when costs and sales are recognized usually varies. This serves as a robustness test to support the main analysis. Formulas for CCC, NTC and their components follow:
13
Payables Deferral Period = ( Accounts Payable / Cost of Goods Sold ) x 365 (3) Cash Conversion Cycle = Inventory Conversion Period + Receivables Conversion Period –
Payables Deferral Period (4)
Net Trade Cycle = ( Accounts Receivable / Total Sales ) x 365 + ( Inventories / Total Sales ) x 365 – ( Accounts Payable / Total Sales ) x 365 (5)
3.2.2 Internationalization as a main independent variable
Firm internationalization dummy variable serves as a primary independent variable. Following Suh, Park, and Yeung (2012), the dummy variable is proxied by the ratio of foreign sales to total sales. Two alternatives are used. For the MNC20 dummy variable, I assign “1” to firms which have at least 20% of foreign sales. For the MNC50 dummy variable, I assign “1” to firms which have at least 50% of foreign sales. The regression equation is:
CCCi,t = α0 + β1 x MNCt-1 + β2 x Sizet-1 + β3 x Growtht-1 + β4 x ROAt-1 +β5 x Debtt-1 + Country
F.E. + Industry F.E. + Year F.E. + εi,t (6)
3.2.3 Country economic development as a moderating variable
For the second hypothesis, I proxy country economic development by aggregate Gross Domestic Product per capita (GDP). I use data provided by The World Bank database reported in US dollars. The research design resembles the methodology of Bruno and Shin (2014), who use interactions of the variables to test for moderating effects. Consequently, I use the interaction of internationalization dummy variable with GDP per capita. The regression equation follows:
CCCi,t = α0 + β1 x MNCt-1 + β2 x GDPt-1 + β3 x MNCt-1 x GDPt-1 + β4 x Sizet-1 + β5 x Growtht-1 +
14
3.2.4 Price stability as a moderating variable
In the third hypothesis, referring to the paper of Goodhart (1995), I proxy price stability by a country-specific inflation rate (Inflation). Inflation is measured as Harmonized Index of Consumer Prices (HICP). Similar as for GDP per capita, I obtain the data from The World Bank. The interaction of the internationalization dummy with inflation is studied.The regression equation for the third hypothesis is as follows:
CCCi,t = α0 + β1 x MNCt-1 + β2 x Inflationt-1 + β3 x MNCt-1 x Inflationt-1 + β4 x Sizet-1 + β5 x
Growtht-1 + β6 x ROAt-1 +β7 x Debtt-1 + Industry F.E. + Year F.E. + εi,t (8) Table 1: Variables
Variable Proxy Shortcut Description Source
WCM efficiency Cash Conversion
Cycle
CCC ( Inventory / Cost of Goods Sold ) x 365 +
( Accounts Receivable / Total Sales ) x 365 - ( Accounts Payable / Cost of Goods Sold ) x 365
Datastream
WCM efficiency Net Trade Cycle NTC ( Accounts Receivable / Total Sales ) x 365 +
( Inventories / Total Sales ) x 365 – ( Accounts Payable / Total Sales ) x 365
Datastream
Internationalization Multinationality
Dummy
MNC20 “1” if foreign sales ratio is higher than 20%; “0”
otherwise
Datastream
Internationalization Multinationality
Dummy
MNC50 “1” if foreign sales ratio is higher than 50%; “0”
otherwise
Datastream Economic
development
Gross Domestic Product per capita
GDP Gross Domestic Product / Population The World Bank
Price stability Inflation Inflation Yearly inflation rate The World Bank
Firm's size Logarithm of
Total Assets
Size ln Total Assets Datastream
Firm's growth Change in Total
Sales
Growth ( Total Salest – Total Salest-1 ) / Total Salest-1 Datastream
Profitability Return on Assets ROA Net Income / Total Assets Datastream
Leverage Debt to Asset
Ratio
Debt Total Debt / Total Assets Datastream
Table 1 summarizes key variables and provides information on proxies, signs, descriptions and data sources used.
3.2.5 Control variables
Following Westerman, Wu, Wiekens, Noordmans, and Laseur (2016), Baños-Caballero, García-Teruel, and Martínez-Solano (2010) and Wasiuzzaman and Arumugam (2013), four additional control variables are used. I control for firm’s size by taking the natural logarithm of Total Assets
15
by Return on Assets (ROA). Debt to Asset Ratio proxies leverage (Debt). For the first set of tests, country, industry and year dummies are employed to control for differences among countries, differences among various industries and developments within the analyzed period. Country-specific effects are left out in the second and third set of tests to avoid multicollinearity. Following formulas are used to compute the control variables:
Total Sales Growth = ( Total Salest – Total Salest-1 ) / Total Salest-1 (9)
Return on Assets = Net Income / Total Assets (10) Debt to Asset Ratio = Total Debt / Total Assets (11)
3.3 Descriptive statistics
Table 2: Descriptive statistics
Quantiles
Variable N Mean Std. Dev. Min 0.25 Median 0.75 Max
CCC 16,474 73.42 63.87 (37.57) 29.74 66.78 108.82 220.13 NTC 16,474 70.17 44.30 2.34 37.46 64.91 95.37 174.33 MNC20 16,474 0.62 0.49 - - 1.00 1.00 1.00 MNC50 16,474 0.36 0.48 - - - 1.00 1.00 GDP 16,474 40,543.65 17,059.62 792.03 36,019.20 46,437.07 49,793.71 119,225.38 Inflation 16,474 0.02 0.02 (0.04) 0.01 0.02 0.03 0.23 Size 16,474 16.30 2.52 12.70 14.35 15.81 17.87 21.56 Growth 16,474 0.09 0.16 (0.20) (0.01) 0.07 0.17 0.46 Debt 16,474 0.24 0.16 - 0.11 0.23 0.35 0.55 ROA 16,474 0.05 0.05 (0.06) 0.02 0.05 0.08 0.17
Table 2 summarizes my key variables used in the main analysis. Total number of observations, mean, standard deviation, and quantiles including minimum, median and maximum are provided. Data are for the period 2006 to 2015.
16
GDP per capita is more than 40,000 US dollars. Inflation reaches on average 2%, which is the value close to the target of the ECB. Considering control variables, values for means of Size, 16.30; Growth, 9%; ROA, 5%; and Debt, 24%, are all sensible. Other characteristics of the variables do not give a reason for further questioning and allow me to continue with the analysis.
3.4. Country-specific descriptive statistics
In Table 3, I provide a country-specific overview of means, medians and standard deviations for the dependent variable CCC. The sample covers a total of 52 countries. Most of the firms are headquartered in the United States of America, Japan and Great Britain. However, the sample is sufficiently diversified and covers enough observations from all over the world.
Morocco shows the highest mean of more than 220 days. Greece and the Philippines have also high values. On the other side of the spectrum are Czechia and Hungary with low CCC. However, to provide a complete picture, it is necessary to mention that some countries with the highest and lowest CCC are represented only by a limited sample of firms. High standard deviations also suggest substantial differences among firms within one country.
17 Table 3: CCC descriptive statistics per country
ISO Mean Median Std. Dev.
Argentina 56.39 23.80 84.77 Australia 57.53 50.50 64.59 Austria 89.66 104.44 56.84 Bahrain 26.59 43.25 44.79 Belgium 75.48 70.67 74.99 Brazil 76.39 65.81 71.10 Canada 43.59 36.54 63.48 Chile 87.24 70.28 83.54 China 92.38 74.35 78.95 Colombia 39.66 37.37 51.93 Czechia (14.72) (19.62) 23.27 Denmark 82.39 74.14 65.23 Egypt 167.25 178.01 53.37 Finland 100.44 96.37 59.40 France 78.94 65.22 76.24 Germany 71.40 81.48 67.15 Greece 142.53 170.22 81.40 Hong Kong 96.13 82.35 82.16 Hungary 3.55 (3.77) 30.58 India 82.93 74.53 72.19 Indonesia 83.30 71.42 59.03 Iraq 27.51 21.50 38.89 Israel 97.21 97.94 63.06 Italy 53.23 41.65 81.03 Japan 79.61 74.62 54.21 Kuwait 43.61 44.21 21.96 Luxembourg 123.53 138.59 80.50 Mexico 53.53 45.08 63.49 Morocco 220.13 220.13 - Myanmar 82.77 70.50 71.08 Netherlands 68.98 72.70 49.75 New Zealand 78.55 61.94 84.10 Norway 89.82 85.44 69.93 Peru 108.53 85.61 57.39 Philippines 116.69 92.16 69.14 Poland 69.64 70.27 59.61 Portugal 41.16 36.91 43.84 Qatar 190.37 192.64 34.57 Russia 85.36 80.12 74.14 Saudi Arabia 81.53 73.25 40.09 Singapore 73.55 49.78 82.23 South Africa 71.05 65.83 54.40 South Korea 75.44 72.12 41.19 Spain 36.08 16.45 72.25 Sri Lanka 80.28 77.19 12.36 Sweden 102.17 99.35 47.58 Switzerland 97.24 98.31 70.84 Thailand 84.11 84.75 51.56 Turkey 121.18 121.15 64.54
United Arab Emirates 80.05 55.09 112.40
United Kingdom of Great Britain and Northern Ireland 70.28 60.77 59.21
United States of America 69.59 64.14 57.36
18
While the detailed descriptive statistics in Table 3 focuses on the main dependent variable, i.e. CCC; I show an overview of a country-specific means on moderating variables, i.e. GDP per capita and inflation, in Table 4. It shows that there are big differences in country economic development among countries and it is also clear that some of the countries are dealing with an increased price instability, as indicated by higher inflation.
Considering GDP per capita, the lowest average value of only 1,282 dollars per capita is allocated to India. Many other Asian countries including China and the Philippines have also very low GDP per capita, indicating that they are far less developed than the average. The average GDP per capita for the whole sample reaches slightly more than 40,000 US dollars. African countries such as Egypt or South Africa; and South American countries such as Argentina and Brazil also do not reach an average. In contrast, European and North American countries have the highest average GDP per capita. Within my sample, Luxembourg is the most developed country. Other highly developed countries include Australia, Switzerland and the United States of America.
19 Table 4: Means of moderating variables per country
ISO GDP per capita Inflation
Argentina 10,968.38 - Australia 55,008.12 0.03 Austria 48,475.82 0.02 Bahrain 23,970.52 0.02 Belgium 45,402.61 0.02 Brazil 10,736.78 0.06 Canada 47,966.76 0.02 Chile 14,443.95 0.03 China 5,310.94 0.03 Colombia 7,095.98 0.03 Czechia 19,688.57 0.02 Denmark 59,686.48 0.02 Egypt 2,657.29 0.11 Finland 48,239.05 0.02 France 41,754.36 0.01 Germany 43,757.63 0.02 Greece 25,783.48 0.02 Hong Kong 34,746.73 0.03 Hungary 13,565.86 0.04 India 1,282.38 0.09 Indonesia 3,131.49 0.06 Iraq 54,012.80 0.01 Israel 30,797.37 0.02 Italy 36,522.39 0.02 Japan 41,165.42 0.00 Kuwait 45,675.25 0.05 Luxembourg 107,295.20 0.02 Mexico 9,639.21 0.04 Morocco 3,160.25 0.00 Myanmar 9,160.29 0.03 Netherlands 51,293.54 0.02 New Zealand 35,749.46 0.02 Norway 91,914.96 0.02 Peru 5,954.86 0.03 Philippines 2,360.94 0.04 Poland 12,861.65 0.03 Portugal 22,403.57 0.02 Qatar 87,417.63 0.03 Russia 12,399.76 0.09 Saudi Arabia 22,258.37 0.04 Singapore 46,852.28 0.03 South Africa 6,571.30 0.06 South Korea 24,240.62 0.02 Spain 30,940.23 0.02 Sri Lanka 2,667.75 0.09 Sweden 54,513.59 0.01 Switzerland 75,771.66 0.00 Thailand 5,148.73 0.03 Turkey 10,851.29 0.08
United Arab Emirates 40,268.56 0.02
United Kingdom of Great Britain and Northern Ireland 43,406.22 0.03
United States of America 49,816.22 0.02
20
3.5. Pearson correlation
To avoid multicollinearity, I check for the correlations among the key variables. Results from a Pearson correlation analysis are tabulated in Table 5. Except for the high correlation between the two dependent variables and the correlation of (0.50) between GDP per capita and inflation, other coefficients do not suggest any problems. As CCC and NTC, and GDP and inflation are not used in one model, this also does not cause any issues for the OLS unbalanced panel regressions.
Table 5: Correlation matrix
Variables CCC NTC MNC20 MNC50 Size Growth Debt ROA GDP Inflation
CCC 1.00 NTC 0.91 1.00 MNC20 0.16 0.18 1.00 MNC50 0.14 0.14 0.59 1.00 Size (0.02) 0.02 0.05 0.01 1.00 Growth 0.01 0.01 (0.06) (0.03) (0.05) 1.00 Debt (0.06) (0.02) (0.03) (0.04) 0.15 (0.05) 1.00 ROA 0.06 0.01 (0.01) 0.01 (0.12) 0.22 (0.36) 1.00 GDP (0.04) (0.07) 0.17 0.12 (0.33) (0.17) (0.04) (0.07) 1.00 Inflation 0.03 0.03 (0.11) (0.05) (0.01) 0.28 (0.00) 0.18 (0.50) 1.00
Table 5 measures correlations among WCM efficiency measures, firm characteristics and country-specific economic characteristics. Data are for the period 2006 to 2015. The sample covers 16,474 observations.
4. Empirical results
4.1 Internationalization and WCM efficiency
I begin with a base analysis of the relationship between internationalization and WCM efficiency. The first set of OLS unbalanced panel regressions for the period 2006 to 2015 includes country, industry and year fixed effects. My results in Table 6 partially support the hypothesis that internationalization has a negative effect on WCM efficiency.
21
shows that this damage to the WCM efficiency is insignificant for firms which are more global and reach at least 50% share of foreign sales.
Table 6: Internationalization and WCM efficiency
Model 1 2 3 4 Dependent variable CCC CCC NTC NTC MNC20 1.98068** 1.69771** (0.95576) (0.67607) MNC50 1.00420 0.24357 (0.85455) (0.60471) Size -1.79503*** -1.75362*** -1.15726*** -1.09722*** (0.59216) (0.59239) (0.41764) (0.41800) Growth -4.38568*** -4.37318*** -5.74006*** -5.71173*** (1.50239) (1.50221) (1.06365) (1.06346) Debt 13.05490*** 13.14580*** 9.47212*** 9.52654*** (2.94161) (2.94173) (2.08079) (2.08102) ROA 19.63557*** 19.64401*** 6.78736 6.71637 (6.14039) (6.14042) (4.34629) (4.34618)
Country F.E. YES YES YES YES
Industry F.E. YES YES YES YES
Year F.E. YES YES YES YES
Constant -90.24275 -89.20382 -19.95329 -19.54518
(63.35068) (63.45275) (44.50809) (44.62713)
Observations 16,474 16,474 16,474 16,474
Countries 52 52 52 52
Table 6 provides results for the OLS panel regression for WCM efficiency. CCC is a dependent variable in Models 1 and 2. NTC is a dependent variable in Models 3 and 4. MNC is a main independent variable. MNC20 is a dummy variable equal to "1" when foreign sales ratio is equal to 20% or higher. MNC50 is a dummy variable equal to "1" when foreign sales ratio is equal to 50% or higher. All models include Size, Growth, ROA and Debt as controls. All models use country, industry and year fixed effects. Period covers years 2006 to 2015.
*** indicate significance at 1% level, ** indicate significance at 5% level and * indicates significance at 10% level. Standard errors are reported in parenthesis.
This is all consistent with the survey of PwC (2018), which shows a slightly negative trend in WCM efficiency over the last couple of years. The previous survey also shows that WCM efficiency of firms was deteriorating (PwC, 2015). Years 2010 and 2014 are the only years covered by my sample for which WCM efficiency improvements are recognized.
22
over the years. Its CCC and NTC also slightly fall with a plummeting share of foreign sales. Indian Tata Motors and Reliance Industry also do not benefit from internationalization. Their CCC and NTC fluctuate close to the mean value.
The literature provides some explanations for my findings. Results are consistent with ideas of Bekaert and Hodrick (2017), who claim that MNCs face several additional obstacles compared to their fellow domestic firms. As they propose, MNCs have higher capital transfer costs and need to deal with fluctuating exchange rates and various tax requirements of foreign tax authorities. If MNCs are inefficient in their strategies on dividend payout policies, transfer pricing, tax planning, hedging and diversification of investments; it can lead to poor WCM efficiency.
My results are also partially consistent with cross-sectional results of Howorth and Westhead (2003), hence larger and older firms may take up fewer WCM routines. However, as Kubick, Lynch, Mayberry, and Omer (2016) propose, even overly aggressive tax and transfer pricing routines may cause some issues. My results show that at least MNCs with the lower share of foreign sales have difficulties with finding the right balance.
Moreover, considering WCM, my results partially contradict the empirical findings of Desai, Foley, and Forbes (2008). At least from the short-term point of view, MNCs cannot benefit from internationalization and fund themselves with international sources of finance. Following other arguments of Bekaert and Hodrick (2017), it also seems like MNCs cannot fully benefit from trade credit provisions, multilateral netting systems and centralized cash management.
23
more and worsen their CCC, but it does not necessary deteriorate a firm’s value. More extreme view is provided by Gill, Biger, and Mathur (2010), who show that even inefficient WCM can lead to better performance. This would explain why firms expand internationally even if it has a negative impact on their CCC and NTC.
Taken together, my findings show no support for Hypothesis 1a and only partial support for Hypothesis 1b. MNCs do not benefit from their international status which turns out to be even harmful for MNCs with a lower share of international sales. Past research papers provide some indicative explanations; however, further empirical research is needed to provide a clear explanation why this happens. It would be interesting to see how various firm strategies and routines moderate the relationship between internationalization and WCM efficiency.
Other results presented in Table 6 show that firm’s size has a positive and significant impact on WCM efficiency, which is in line with findings of Baños-Caballero, García-Teruel, and Martínez-Solano (2010). Negative impact of profitability is also consistent with the results of the same authors. Against their findings, I find a positive effect of sales growth on WCM efficiency. Consistent with Wasiuzzaman and Arumugam (2013), I find that higher leverage prolongs the CCC. Jointly significant results for country as well as industry dummies are not tabulated but show the importance of controlling for these effects as there are significant differences among countries and industries. This is consistent with the previous idea of Preve and Sarria-Allende (2010).
4.2 The role of country economic development
24
the CCC, or alternatively NTC. However, the role of GDP per capita is visible once I apply it as a moderator. It turns the signs of MNC dummies. Results are significant in Models 3, 7 and 8.
Table 7: Country economic development and WCM efficiency
Model 1 2 3 4 Dependent variable CCC CCC CCC CCC MNC20 1.48350 -4.83345** (0.94694) (1.98601) MNC50 0.87144 -1.79252 (0.84553) (2.07447) GDP 0.00002 0.00002 -0.00009 -0.00001 (0.00005) (0.00005) (0.00006) (0.00005) MNC20 x GDP 0.00017*** (0.00005) MNC50 x GDP 0.00007 (0.00005) Size 0.19456 0.20259 0.17010 0.18532 (0.41898) (0.41905) (0.41790) (0.41825) Growth -4.51637*** -4.50669*** -4.35128*** -4.46534*** (1.50314) (1.50317) (1.50408) (1.50416) Debt 11.56746*** 11.64680*** 11.63396*** 11.67786*** (2.92202) (2.92167) (2.92139) (2.92214) ROA 21.90656*** 21.92235*** 21.20580*** 21.77918*** (6.13563) (6.13631) (6.13922) (6.13964) Country F.E. NO NO NO NO
Industry F.E. YES YES YES YES
Year F.E. YES YES YES YES
Constant -47.25469 -45.99641 -42.32965 -45.06259
(60.89291) (60.90880) (60.70120) (60.73968)
Observations 16,474 16,474 16,474 16,474
Countries 52 52 52 52
25 Table 7: Continued Model 5 6 7 8 Dependent variable NTC NTC NTC NTC MNC20 1.38170** -3.72079*** (0.67066) (1.40659) MNC50 0.24240 -4.26175*** (0.59896) (1.46933) GDP -0.00002 -0.00001 -0.00010*** -0.00006* (0.00003) (0.00003) (0.00004) (0.00004) MNC20 x GDP 0.00014*** (0.00003) MNC50 x GDP 0.00011*** (0.00003) Size 0.40598 0.41651 0.38602 0.39043 (0.29603) (0.29616) (0.29499) (0.29519) Growth -5.89163*** -5.86450*** -5.75683*** -5.79330*** (1.06507) (1.06513) (1.06581) (1.06580) Debt 8.53416*** 8.60907*** 8.59469*** 8.68507*** (2.06957) (2.06947) (2.06895) (2.06932) ROA 8.43437* 8.37496* 7.86238* 8.11472* (4.34702) (4.34771) (4.34971) (4.34971) Country F.E. NO NO NO NO
Industry F.E. YES YES YES YES
Year F.E. YES YES YES YES
Constant 14.46676 15.62969 18.45443 17.12857
(42.99286) (43.01783) (42.80657) (42.82624)
Observations 16,474 16,474 16,474 16,474
Countries 52 52 52 52
Table 7 provides results for the OLS panel regression for WCM efficiency, when internationalization dummy is interacted with country economic development. CCC is a dependent variable in Models 1-4. NTC is a dependent variable in Models 5-8. MNC is a main independent variable. MNC20 is a dummy variable equal to "1" when foreign sales ratio is equal to 20% or higher. MNC50 is a dummy variable equal to "1" when foreign sales ratio is equal to 50% or higher. GDP, i.e. moderating variable, stands for GDP/ capita. In models 3, 4, 7 and 8, MNC dummy is interacted with GDP variable. All models include Size, Growth, ROA and Debt as controls. All models use industry and year fixed effects. Period covers years 2006 to 2015.
*** indicate significance at 1% level, ** indicate significance at 5% level and * indicates significance at 10% level. Standard errors are reported in parenthesis.
26
days for the largest firms. This result is significant on 1% level. Other models with different dependent variables or different MNC dummies support robustness of my results.
PwC (2015) survey shows that Asian firms from developing countries have generally the worst WCM efficiency. However, my examples show that Asian MNCs can even outperform comparable MNCs from developed countries. The case from automotive industry supports my results. German Bayerische Motoren Werke has less efficient WCM than Indian Tata Motors across all the observed years. Another example from electronics industry tells the same story. Chinese Lenovo seems to be much more efficient than South Korean Samsung Electronics.
As hypothesized, my results provide an evidence that country economic development has a negative effect on the relationship between internationalization and working capital management efficiency. Thus, there is full support for my second hypothesis. Despite of the claim of Healy and Palepu (2001), WCM efficiency does not seem to be related to disclosures and transparency which is higher in developed markets. It also looks like MNCs from developing countries do not suffer from poor managerial skills as it was generalized by Agyei-Mensah (2010). However, to empirically support these interpretations, further research is recommended.
4.3 The role of price stability
27
Contingent on the dependent variable, coefficients of interactions have even opposite signs and very high standard deviations.
Table 8: Inflation and WCM efficiency
Model 1 2 3 4 Dependent variable CCC CCC CCC CCC MNC20 1.47716 1.26902 (0.94606) (1.16410) MNC50 0.83386 0.54786 (0.84533) (1.07201) Inflation 58.85831*** 58.72383*** 54.15893** 54.86664** (20.32234) (20.32491) (25.29544) (22.17803) MNC20 x Inflation 8.37130 (26.53971) MNC50 x Inflation 12.36636 (28.13455) Size 0.10207 0.10538 0.09923 0.09846 (0.40812) (0.40821) (0.40783) (0.40817) Growth -4.99215*** -4.98575*** -5.00708*** -4.99511*** (1.50596) (1.50603) (1.50708) (1.50648) Debt 11.63123*** 11.71770*** 11.61440*** 11.69445*** (2.91716) (2.91672) (2.91745) (2.91710) ROA 21.76507*** 21.76798*** 21.73588*** 21.70668*** (6.13071) (6.13143) (6.13293) (6.13433) Country F.E. NO NO NO NO
Industry F.E. YES YES YES YES
Year F.E. YES YES YES YES
Constant -46.61534 -45.20706 -46.46215 -44.94757
(60.86164) (60.87219) (60.79824) (60.81921)
Observations 16,474 16,474 16,474 16,474
Countries 52 52 52 52
28 Table 8: Continued Model 5 6 7 8 Dependent variable NTC NTC NTC NTC MNC20 1.34976** 1.71580** (0.66999) (0.82447) MNC50 0.20580 0.67383 (0.59875) (0.75935) Inflation 43.98188*** 44.05854*** 52.05203*** 50.35224*** (14.39379) (14.39649) (17.91657) (15.70909) MNC20 x Inflation -14.13643 (18.80022) MNC50 x Inflation -19.83317 (19.93101) Size 0.40177 0.40684 0.40532 0.41696 (0.28859) (0.28871) (0.28833) (0.28863) Growth -6.17705*** -6.15570*** -6.15013*** -6.13932*** (1.06682) (1.06691) (1.06763) (1.06724) Debt 8.48216*** 8.56341*** 8.48794*** 8.58260*** (2.06599) (2.06582) (2.06618) (2.06605) ROA 8.44420* 8.37278* 8.51020* 8.48517* (4.34268) (4.34341) (4.34431) (4.34548) Country F.E. NO NO NO NO
Industry F.E. YES YES YES YES
Year F.E. YES YES YES YES
Constant 12.95345 14.25301 12.70675 13.85811
(43.02098) (43.03918) (42.96662) (42.99208)
Observations 16,474 16,474 16,474 16,474
Countries 52 52 52 52
Table 8 provides results for the OLS panel regression for WCM efficiency, when internationalization dummy is interacted with price stability. CCC is a dependent variable in Models 1-4. NTC is a dependent variable in Models 5-8. MNC is a main independent variable. MNC20 is a dummy variable equal to "1" when foreign sales ratio is equal to 20% or higher. MNC50 is a dummy variable equal to "1" when foreign sales ratio is equal to 50% or higher. Inflation is a proxy for price stability. In models 3, 4, 7 and 8, MNC dummy is interacted with Inflation variable. All models include Size, Growth, ROA and Debt as controls. All models use industry and year fixed effects. Period covers years 2006 to 2015.
*** indicate significance at 1% level, ** indicate significance at 5% level and * indicates significance at 10% level. Standard errors are reported in parenthesis.
29
According to De Grauwe (2018), inflation impacts purchasing power, country debt, growth and even employment. My results show that these issues are the equal problem for international and local firms.
4.4 Sensitivity analysis
Building on my previous results for Hypothesis 2 and 3, I examine the least economically developed countries with GDP per capita under 10,000 US dollars and countries with an inflation rate above 2% in the further sensitivity analysis. I want to show that my results are valid for countries with a low development status and price instability. Table 9 presents the brief descriptive statistics for my sensitivity analysis. All the coefficients are meaningful and relatively comparable with the values for the whole sample presented in Table 2.
Table 9: Descriptive statistics for the sensitivity analysis
Variable N Mean Std. Dev. Min Max
Economically underdeveloped countries
CCC 1,957 82.91 69.92 (37.57) 220.13 NTC 1,957 79.02 52.35 2.34 174.33 MNC20 1,957 0.38 0.49 - 1.00 Size 1,957 17.59 1.98 12.79 21.56 Growth 1,957 0.16 0.17 (0.20) 0.46 Debt 1,957 0.24 0.16 - 0.55 ROA 1,957 0.07 0.06 (0.06) 0.17 Price instability CCC 7,851 74.47 65.06 (37.57) 220.13 NTC 7,851 70.82 45.56 2.34 174.33 MNC20 7,851 0.59 0.49 - 1.00 Size 7,851 16.09 2.35 12.70 21.56 Growth 7,851 0.12 0.16 (0.20) 0.46 Debt 7,851 0.23 0.16 - 0.55 ROA 7,851 0.06 0.05 (0.06) 0.17
Table 9 summarizes my key variables used in the sensitivity analysis. Total number of observations, mean, standard deviation, minimum and maximum are provided. Data are for the period 2006 to 2015.
4.4.1 Economically underdeveloped countries
30
observations and covers countries from various continents. Mainly firms from China, Colombia, Egypt, India, Indonesia, Mexico, Morocco, Myanmar, Peru, Philippines, South Africa, Sri Lanka and Thailand are included in the sample.
Coefficients for an MNC20 dummy in Table 10 turn to negative numbers. This fully supports my previous results from Tables 6 and 7. It is clear from Models 1 and 2 that being an MNC decreases the CCC and the NTC and consequently increases WCM efficiency. These results are both economically and statistically significant.
Table 10: Internationalization and WCM efficiency for less developed countries
Model 1 2 Dependent variable CCC NTC MNC20 -6.03089** -3.89775* (3.03949) (2.15995) Size 0.70625 2.54848* (1.93116) (1.39138) Growth -7.81679 -7.16437** (5.05604) (3.58034) Debt 30.39561*** 12.28269* (10.39027) (7.38684) ROA 57.23681** 5.62497 (25.59424) (18.15379)
Country F.E. YES YES
Industry F.E. YES YES
Year F.E. YES YES
Constant 176.92608** 112.62546*
(78.93546) (58.37739)
Observations 1,957 1,957
Table 10 provides results for the OLS panel regression for WCM efficiency. CCC is a dependent variable in Model 1. NTC is a dependent variable in Model 2. MNC is a dummy variable equal to "1" when foreign sales ratio is equal to 20% or higher. Both models include Size, Growth, ROA and Debt as controls. All models use country, industry and year fixed effects. Period covers years 2006 to 2015. Sample consists of countries with GDP per capita lower than 10,000 dollars.
*** indicate significance at 1% level, ** indicate significance at 5% level and * indicates significance at 10% level. Standard errors are reported in parenthesis.
4.4.2 Price instability
31
sample are Egypt, India, Indonesia, Russia, Sri Lanka and Turkey. The sample for this test consists of 7,851 observations.
Table 11: Internationalization and WCM efficiency for high inflation countries
Model 1 2 Dependent variable CCC NTC MNC20 3.66994*** 2.59482** (1.41354) (1.01843) Size -1.74438** -0.69307 (0.75404) (0.53327) Growth -5.00505** -4.82078*** (2.24181) (1.62836) Debt 9.95285** 6.96522** (4.51289) (3.24943) ROA 22.19294** 4.07796 (9.79421) (7.09521)
Country F.E. YES YES
Industry F.E. YES YES
Year F.E. YES YES
Constant -92.47969 -33.08062
(64.21118) (44.92627)
Observations 7,851 7,851
Table 11 provides results for the OLS panel regression for WCM efficiency. CCC is a dependent variable in Model 1. NTC is a dependent variable in Model 2. MNC is a dummy variable equal to "1" when foreign sales ratio is equal to 20% or higher. Both models include Size, Growth, ROA and Debt as controls. All models use country, industry and year fixed effects. Period covers years 2006 to 2015. Sample consists of countries with inflation above 2%.
*** indicate significance at 1% level, ** indicate significance at 5% level and * indicates significance at 10% level. Standard errors are reported in parenthesis.
32
5. Conclusion
5.1 Conclusions and contributions
In this research paper, I have studied the relationship between internationalization and WCM efficiency. I have applied country economic development and price stability as moderating variables. I see my main contribution in an explanation of the impact of internationalization on WCM efficiency and providing a picture on the role of aforementioned macroeconomic factors.
My empirical findings link the world of international finance, operations management, and the macroeconomic literature. While most of the previous literature studies long-term capital management, my results show the importance of short-term capital management through its determinants. My findings illustrate the differences in WCM efficiency between MNCs and domestic firms. Taken together, my results do not show an advantage of MNCs over domestic firms in good economic conditions. I can even conclude that domestic firms are more efficient in such a situation. However, the advantage of MNCs becomes statistically and economically significant when they are headquartered in less developed countries. High inflation has a negative impact on firms; however, there is no significant difference between MNCs and local firms.
33
5.2 Managerial implications
My empirical findings have several implications for managers of MNCs. In general, MNCs have longer CCC and thus, less efficient WCM. However, as Aktas, Croci, and Petmezas (2015) claim, there might be an optimal level of WCM. My results show that this level might differ for MNCs and domestic firms. The key implication of my research paper is that managers of MNCs should pay more attention to WCM and try to shorten their CCC. PwC (2018) reports that it is not always the case over the last couple of years.
Managers from less developed and high inflation countries should focus on WCM efficiency and benefit from advantages offered by foreign markets. Obtaining easy short-term funding, lower interest rates on short-term loans, higher interest rates on short-term deposits, various trade credit arrangements with customers and suppliers or other firm-specific economies of scale can be highly beneficial for managers of these MNCs and improve WCM efficiency.
Taking a different point of view leads to my last implication for managers of MNCs. Either when acquiring a foreign firm or when a firm is being acquired by a foreign entity, internationalization clearly plays a role in valuation of WCM policies. This can significantly influence the acquisition value and thus requires superior management skills. Depending on which side managers take in an acquisition process, proper WCM can either increase or decrease price of an acquisition.
5.3 Limitations and future research
34
future research could study the effect of internationalization on inventory conversion period, receivables conversion period and payables deferral period.
Second, this research does not answer the question whether there is an optimal level of WCM for firms as suggested by Aktas, Croci, and Petmezas (2015). It also does not study whether an efficient WCM of MNCs is beneficial for shareholders in terms of profitability. These may be further questions to answer in the future research. Aktas, Croci, and Petmezas (2015) provide solutions to most of the issues within the WCM field; however, my research shows that it is still worth to study some specific aspects which may influence WCM efficiency.
Third, more research is necessary to understand why MNCs have, in general, less efficient WCM than local firms. Considering various strategies and routines of MNCs, which are in line with Howorth and Westhead (2003) and Kubick, Lynch, Mayberry, and Omer (2016), could provide an interesting explanation.
Fourth, apart from tabulated variations among countries, results also reveal extensive differences among industries. I control for these effects, but I do not provide results of these tests in my tables. Anyway, it would be interesting to study how internationalization impacts WCM efficiency across various industries and what are the drivers for the efficiency within specific industries. However, the researcher must necessarily distinguish between what are the country-specific effects and what are the industry-specific effects in this kind of research. Some countries may specialize in a certain industry and this can cause a bias.
6. References
35
Aktas, N., Croci, E., Petmezas, D., 2015. Is Working Capital Management Value-Enhancing? Evidence from Firm Performance and Investments. Journal of Corporate Finance 30, 98–113.
Bekaert, G., Hodrick, R.J., 2017. International Financial Management. Pearson, London.
Baños-Caballero, S., García-Teruel, P.J., Martínez-Solano, P., 2010. Working Capital Management, Corporate Performance, and Financial Constraints. Journal of Business Research 67, 332–338.
Boschker, B.A., 2011. Determinants of Working Capital Management in SMEs. Tilburg University, Tilburg.
Bruno, V., Shin, H.S., 2014. Globalization of Corporate Risk Taking. Journal of International Business Studies 45, 800–820.
Chang, C.-C., 2018. Cash Conversion Cycle and Corporate Performance: Global Evidence. International Review of Economics & Finance 56, 568–581.
Chang, K., Noorbakhsh, A., 2009. Does National Culture Affect International Corporate Cash Holdings? Journal of Multinational Financial Management 19, 323–342.
Cuñat, V., 2008. Trade Credit: Suppliers as Debt Collectors and Insurance Providers. Review of Financial Studies 20, 491–527.
Desai, M., Foley, C.F., Forbes, K., 2008. Financial Constraints and Growth: Multinational and Local Firm Responses to Currency Crises. Review of Financial Studies 21, 2857–2888.
36
Ferreira, M.A., Vilela, A.S., 2004. Why Do Firms Hold Cash? Evidence from EMU Countries. European Financial Management 10, 295–319.
Gill, A., Biger, N., Mathur, N., 2010. The Relationship Between Working Capital Management and Profitability: Evidence from The United States. Business and Economics Journal 10, 1–9.
Goodhart, C.A.E., 1995. Price Stability and Financial Fragility. Financial Stability in a Changing Environment 439–509.
Grauwe, P.D., 2018. Economics of Monetary Union. Oxford University Press, Oxford.
Haan, J.D., Vlahu, R., 2016. Corporate Governance of Banks: A Survey. Journal of Economic Surveys 30, 228–277.
Healy, P.M., Palepu, K., 2001. Information Asymmetry, Corporate Disclosure and the Capital Markets: A Review of the Empirical Disclosure Literature. Journal of Accounting and Economics 31, 405–440.
Howorth, C., Westhead, P., 2003. The Focus of Working Capital Management in UK Small Firms. Management Accounting Research 14, 94–111.
Jensen, M.C., 1986. Agency Cost of Free Cash Flow, Corporate Finance, and Takeovers. American Economic Review 323–329.
Kubick, T.R., Lynch, D., Mayberry, M.A., Omer, T.C., 2016. The Effects of Regulatory Scrutiny on Tax Avoidance: An Examination of SEC Comment Letters. The Accounting Review 91, 1751– 1780.
37
Madura, J., Whyte, A.M., 1990. Diversification Benefits of Direct Foreign Investment. Management International Review 30, 73–85.
Martínez-Solano, P., García-Teruel, P.J., 2007. Effects of Working Capital Management on SME Profitability. International Journal of Managerial Finance.
Mongrut, S., O’Shee, D.F., Zavaleta, C.C., Zavaleta, J.C., 2012. Determinants of Working Capital Management in Latin American Companies. Innovar 24, 5–17.
Moyer, R.C., McGuigan, J.R., Rao, R.K.S., 2017. Contemporary Financial Management. Cengage Learning, Boston, USA.
Platt, H., 2010. Lead with Cash: Cash Flow for Corporate Renewal. Imperial College Press.
Preve, L., Sarria-Allende, V., 2010. Working Capital Management. Oxford: Oxford University Press.
PwC, 2015. Bridging the Gap: 2015 Annual Global Working Capital Survey 1–52.
PwC, 2018. Navigating Uncertainty: PwC’s Annual Global Working Capital Study 1–32.
Reeb, D.M., Kwok, C.C.Y., Baek, H.Y., 1998. Systematic Risk of the Multinational Corporation. Journal of International Business Studies 29, 263–279.
Richards, V.D., Laughlin, E.J., 1980. A Cash Conversion Cycle Approach to Liquidity Analysis. Journal of Financial Management 9, 32.
38
Shin, H., Soenen, L., 1998. Efficiency of Working Capital Management and Corporate Profitability. Financial Practice & Education 28, 39–45.
Smith, J.K., 1987. Trade Credit and Informational Asymmetry. The Journal of Finance 42, 863– 872.
Suh, J., Park, S., Yeung, B.Y., 2012. Do Multinational Corporations and Domestic Corporations Differ in Leverage Policies? Journal of Corporate Finance 20, 115–139.
Tewolde, S., Eije, H.V., 2007. Selfishness or Co-operation: Trade Credit in Manufacturing Firms in Eritrea. Accounting, Banking and Corporate Financial Management in Emerging Economies Research in Accounting in Emerging Economies 369–387.
Wasiuzzaman, S., Arumugam, V.C., 2013. Determinants of Working Capital Investment: A Study of Malaysian Public Listed Firms. Australasian Accounting, Business and Finance Journal 7, 63– 83.
Westerman, W., 2015. Working Capital Management Programs: Yesterday, Today, Tomorrow. Journal of Modern Accounting and Auditing 11.
