INTRODUCTION
INTRODUCTION AND BACKGROUND
Business managers aim to create value for their companies and shareholders by effectively utilizing available resources, particularly operational cash flow Cash flow volatility, defined as the fluctuation in a firm's operating cash flow over time, significantly impacts corporate investment, which encompasses capital expenditures, research and development, acquisitions, and advertising costs Operational cash flow is essential for shareholder value creation, as it provides liquidity for paying dividends and funding investment activities Fluctuations in cash flow can lead to shortages that adversely affect investment activities, as cash flow serves as a critical internal funding source Consequently, corporate investment may decline during periods of cash flow volatility, negatively impacting firm value Investment is vital for achieving corporate goals, and profitable projects enhance earnings and market value Furthermore, investing in high-growth opportunities can stabilize cash flow volatility Research indicates a significant relationship between cash flow volatility, corporate investment, and firm value, with corporate investment potentially acting as a mediating factor in this relationship, underscoring its importance to business operations and overall firm value.
Figure 1.1 The expected relationship between cashflow volatility, corporate investment and firm value.
Numerous studies over the years have explored the complex relationships between cash flow volatility, corporate investment, and firm value, yet the findings remain contentious A significant body of research indicates a negative correlation between cash flow volatility and corporate investment, as evidenced by the works of Minton and Shrand (1999), Beladi et al (2021), Panagiotidis and Printzis (2021), and Rashid et al (2021) In contrast, alternative perspectives, such as those presented by Baum et al (2010) and Cohen, suggest differing conclusions, highlighting the ongoing debate in this area of study.
Research by Kimaiyo (2017) and Chortareas et al (2021) highlights a positive relationship between cash flow volatility and firm value, aligning with findings from 2014 However, contrasting studies by Minton et al (2002), Rountree et al (2008), Mọkelọ (2012), and Altuntas et al (2017) indicate a negative impact of cash flow volatility on firm value Previous research by Shipe (2015) and Chi and Su further explores these dynamics.
Research by Gworo (2019) and Sawalqa (2021) indicates that fluctuations in cash flow positively influence company value Multiple studies, including those by Johnson and Pazderka (1993), Ye and Yuan (2008), Kim et al (2018), Dang et al (2020), and Mousa et al (2021), have established a positive correlation between corporate investment and firm value However, most previous research has examined these factors in isolation, with limited studies, like Njuguna et al (2022), analyzing the combined effects of these variables Njuguna et al utilized Baron and Kenny's (1986) four-step mediating model to explore how corporate investment impacts firm value amid cash flow fluctuations, starting by identifying a significant relationship between independent and dependent variables, followed by assessing the influence of the independent variable on the mediating variable, then examining the connection between the mediating and dependent variables.
3 fourth step focuses on detecting the mediating effect by studying simultaneously the impacts of the independent and the mediating variable on the dependent variable
Further investigation into the interplay between cash flow volatility, corporate investment, and firm value is essential, as it could yield valuable empirical insights and a more comprehensive understanding of the relationships among these key financial variables.
Vietnam, as a developing Asian nation with a market-oriented economy, faces economic fluctuations and potential imbalances (Tran & Le, 2017) Vietnamese enterprises encounter significant uncertainty from various factors, including market conditions and internal challenges The Covid-19 pandemic has exacerbated these insecurities, leading to supply chain disruptions and lockdowns that have severely impacted business operations Consequently, companies have experienced fluctuations in operating cash flow, which have diminished their liquidity and hindered their ability to invest in business activities.
During this period, the Government implemented policies to increase money supply and lower interest rates to stimulate enterprise investment However, these measures resulted in inflation, prompting the State to raise interest rates, which created challenges for many Vietnamese businesses in securing loans due to insufficient internal funding Consequently, this led to a decline in investment activities and a decrease in business value The importance of studying cash flow fluctuations, corporate investment, and their impact on firm value has become evident, as such research can help mitigate operational risks and enhance company value in Vietnam Currently, there are limited domestic studies on this topic, with notable research indicating that fixed asset investment positively influences firm performance, while cash flow negatively affects investment activities, particularly for confident CEOs Additionally, evidence suggests that overinvestment adversely impacts corporate performance, and firms with excess free cash flow are prone to overinvestment The existing literature primarily focuses on cash flow's impact, with few studies addressing cash flow fluctuations or the simultaneous relationship between cash flow volatility, corporate investment, and firm value, highlighting the need for further exploration in these areas for Vietnamese enterprises.
4 enterprises to pay more attention to investment management and operating cash flow control to minimize risk and maximize business value
This study explores the mediating effect of corporate investment on the relationship between cash flow volatility and firm value among listed non-financial firms on the Ho Chi Minh Stock Exchange Driven by a curiosity about how these factors interact, the author has chosen this topic for their dissertation The research will build on the scientific foundation established by Njuguna and colleagues in their 2022 study.
RESEARCH GAP IDENTIFICATION AND NEW CONTRIBUTIONS
The author identifies significant gaps in existing research regarding the interplay between cash flow volatility, corporate investment, and firm value To address these gaps, the study presents four key contributions aimed at enhancing the understanding of these variables and their relationships in the context of corporate finance.
Research on the relationship between cashflow volatility and firm performance in Vietnam is limited, with existing studies highlighting various aspects of investment and performance For instance, Dang et al (2020) demonstrated that fixed asset investment positively influences firm performance, while Nguyen (2021) identified a statistically significant negative effect of overinvestment on corporate performance Additionally, Linh (2022) found that firms with excess free cashflows are prone to overinvestment However, these studies primarily concentrate on cashflow rather than cashflow volatility This study aims to fill that gap by providing empirical evidence on the interactions between cashflow volatility, corporate investment, and firm value in the context of a developing country like Vietnam.
Research on the mediating role of corporate investment in the relationship between cash flow fluctuations and enterprise value is scarce globally, with no studies conducted in Vietnam Existing literature on these three variables presents inconsistencies This study addresses this gap by examining the mediating effect of corporate investment on the relationship between cash flow volatility and firm value for non-financial firms listed on HOSE from 2018 to 2022, analyzing all three variables concurrently This approach contrasts with previous studies that focused solely on the individual relationships among cash flow volatility, corporate investment, and firm value, thereby offering a comprehensive understanding of how cash flow fluctuations influence corporate value.
Third, the dissertation gives an updated analysis of the relationships of these variables in Vietnam with a fresh set of data including 1,500 firm-year observations of
300 non-financial listed firms on Ho Chi Minh Stock Exchange (HOSE), during the period 2018-2022 These figures are highly topical because they are close to the current time
This dissertation enhances research methodology by employing the four-step model of Baron and Kenny (1986) to examine mediating variables, utilizing innovative approaches like FEM, REM with clustered standard errors, and System-GMM It highlights the need for more comprehensive studies on this topic in Vietnam, as there is a scarcity of official Vietnamese research addressing mediating variables.
RESEARCH OBJECTIVES
This dissertation aims to explore how corporate investment mediates the relationship between cash flow volatility and the value of non-financial firms listed on HOSE from 2018 to 2022, utilizing the four-step mediating research model proposed by Baron and Kenny (1986) The research will systematically progress through these four essential steps, illustrated in Figure 1.1.
Figure 1.2 Research objectives of the dissertation based on the 4-step model of
To establish the mediating effect of corporate investment on the relationship between cash flow volatility and firm value, it is essential to first confirm a significant impact of cash flow volatility on firm value Step 1 focuses on this crucial relationship Following this, Steps 2 and 3 examine how cash flow volatility influences corporate investment and how corporate investment, in turn, affects firm value The core of this analysis, Step 4, aims to determine whether corporate investment mediates the relationship between cash flow volatility and firm value However, a definitive conclusion can only be reached if a significant link among these three variables is established, along with the fulfillment of specific conditions regarding the regression coefficients.
RESEARCH QUESTIONS
In order to achieve the research objective, there are 1 key research question, and
3 particular research questions related to each step in the research model as below:
• Key question: Does corporate investment have a mediating effect on the link between cashflow volatility and firm value of non-financial firms listed on HOSE?
• Sub-question 1: Does cashflow volatility have a significant impact on firm value?
• Sub-question 2: Does cashflow volatility have a significant impact on corporate investment?
• Sub-question 3: Does corporate investment have a significant impact on firm value?
THE SCOPE OF THE STUDY
The dissertation will investigate 300 non-financial firms listed on HOSE between
The author analyzes Vietnam's economic landscape from 2018 to 2022 to understand the impacts of the Covid-19 pandemic, which introduced significant uncertainties across social, economic, and political spheres This five-year period is crucial for examining how businesses navigated investment decisions amid fluctuating operations and cash flow, with many facing the risk of bankruptcy The availability of accessible data enhances the study's reliability, while the exclusion of financial firms ensures unbiased results, given their distinct operational characteristics compared to non-financial firms.
RESEARCH DATA AND METHODOLOGY
This research utilizes secondary data from the Fiinpro X database, focusing on 300 non-financial firms listed on the HOSE from 2013 to 2022, analyzing key metrics such as operating cash flows, capital expenditure, depreciation, and market value of equity Financial firms, including banks and insurance companies, are excluded due to their unique capital structures and regulatory frameworks Additionally, certain inappropriate firms were omitted, resulting in a total of 1,500 firm-year observations, though this number may fluctuate based on specific circumstances The sample represents 77.92% of the companies and approximately 70% of the total market capitalization of non-financial firms on the HOSE.
This study will investigate the mediating effect using the research model established by Baron and Kenny (1986), as referenced in Njuguna et al (2022) The initial step focuses on identifying a significant relationship between the independent and dependent variables.
In the second step, it is crucial to establish that the independent variable has a significant impact on the mediating variable The third step aims to identify a meaningful relationship between the mediating variable and the dependent variable Lastly, the fourth step involves examining the mediating effect by analyzing the simultaneous influences of both the independent and mediating variables on the dependent variable.
Previous studies have utilized cross-sectional regression to explore the relationship between cash flow volatility and firm value, but this method often neglects the heterogeneity that can distort results To address this issue, Njuguna et al (2022) and Kimaiyo (2017) employed panel regression, which combines cross-sectional and time series data, enhancing degrees of freedom and reducing multi-collinearity among predictor variables Kimaiyo’s research specifically applied the Pool-Ordinary Least Square (POLS) model to analyze the effects of cash flow volatility and corporate investment; however, this approach has limitations, such as potential misinterpretation of coefficients due to collinearity and the necessity for more observations than independent variables Consequently, this dissertation will build upon previous methodologies by adopting a panel regression model, incorporating Fixed and Random-Effects Models (FEM and REM) to implement the four-step mediation model proposed by Baron and Kenny (1986), with FEM highlighting the distinct characteristics of individual units.
In scenarios where individual-specific intercepts may correlate with one or more regressors, the Fixed Effects Model (FEM) is appropriate, though it can consume significant degrees of freedom with large cross-sectional units Conversely, the Random Effects Model (REM) assumes that the intercept for each unit is randomly drawn from a larger population with a constant mean, making it suitable when these intercepts are uncorrelated with the regressors To determine the most consistent model between FEM and REM, the Hausman test will be employed, followed by diagnostic tests to identify any issues within each research model Ultimately, the System-Generalized Method of Moments (System-GMM) developed by Blundell and Bond, along with REM utilizing clustered standard errors, will be implemented to address model-related problems and draw final conclusions.
RESEARCH STRUCTURE
The overall structure of the dissertation takes the forms of 5 chapters, which has been presented in the following way
Chapter 1 (introduction) briefly presents the overall background, research gaps, and reasons leading to the research Then, it indicates the research objective and related questions, research data, and methodology
Chapter 2 (Literature review) presents the framework theory, literature review, definition and calculation of variables used in the study, and some previous research Finally, it indicates the hypotheses of the research model
Chapter 3 (Research methodologies) has three objectives First, it presents information about the data, samples, and variables used in the study Second, it shows the models used for the research question Third, it describes the research methods and procedures that are applied throughout this dissertation
Chapter 4 (Empirical results) depicts the results of models through each method and gives the analysis, evaluation, and conclusions
Chapter 5 (Conclusion and Policy Implications) gives the findings, recommendations, and limitations
LITERATURE REVIEW
THEORETICAL FRAMEWORK
Pecking order theory elucidates the impact of cash flow volatility on investment activities by highlighting the preference for utilizing free cash flow within a company When a business experiences abundant cash flow, it tends to increase its investment activities; however, in times of scarce operating cash flow, firms may curtail investments due to the high costs associated with external financing.
Myers and Majluf (1984) highlighted that firms' capital structure decisions are influenced by information asymmetries, where business leaders possess the most insight into their company's internal circumstances, including growth potential and challenges To address this disparity, external funding sources demand higher returns to offset the associated risks, resulting in increased funding costs Consequently, managers prioritize capital sources in a flexible manner, typically opting for internal funds first, followed by debt, and lastly new equity.
(1984), Hillier et all (2010), and Myers (2001) further reinforced the existence of this hierarchy of priorities in making financial decisions
(Source: Kaplan Financial Knowledge Bank, 2013, author's synthesis)
Li et al (2011) highlight that when selecting financing for new projects, companies prioritize retained earnings over debt, short-term debt over long-term debt, and debt over equity Internal funding sources like retained earnings are favored due to their availability and lower costs associated with asymmetric information (Gaud et al., 2005; Mazur, 2007; Mostafa & Boregowda, 2014) Luigi and Sorin (2009) further affirm that internal funding is the most convenient and cost-effective option Conversely, external financing methods, such as new debt or equity issuance, incur higher agency costs for companies Debt holders typically demand a lower return than stockholders because they have a stronger claim on capital in bankruptcy scenarios This increase in capital costs can negatively impact a company's value, as noted by Jensen and Meckling (1976).
The underinvestment theory is crucial to this research as it highlights the connection between investment activity and firm value, which are key variables in the dissertation This theory identifies three financing methods—internal funding, debt, and equity issuance—that contribute to underinvestment Consequently, this behavior leads to diminished investment activities, ultimately resulting in a decrease in the enterprise's value.
Underinvestment occurs when a company consistently overlooks opportunities to invest in projects with a positive net present value, ultimately hindering its market value This concept, introduced by Myers in 1977, highlights the financial dynamics between shareholders and creditors in leveraged firms, where shareholders may forgo potential investments due to concerns that creditors will capture the majority of the resulting profits.
Companies with high debt ratios prioritize shareholder interests differently than those with low or no debt, as noted by Myers (1977) To enhance company value, managers typically pursue projects with positive NPVs However, shareholders of high-risk debt firms are often reluctant to finance new projects using internal funds, as profits are primarily allocated to debt repayment Stein (2001) highlights that risky debt functions as a "tax" on new investments, diverting potential value away from growth initiatives This dynamic can lead managers to forgo valuable projects, ultimately resulting in a decline in overall business value.
Underinvestment in debt-financed projects, as explained by Jensen and Meckling (1976), arises from the ethical dilemma of asset substitution Managers may exploit low borrowing costs to fund high-risk projects while misleading creditors by using investment loans intended for low-risk ventures In response, creditors often raise interest rates or impose strict bond covenants to mitigate this risk However, shareholders perceive these funding costs as excessively high, leading them to reject potentially lucrative projects, ultimately resulting in underinvestment.
Another way to finance a project is by issuing new equity, and Myers and Majluf
In 1984, it was highlighted that underinvestment occurs due to asymmetric information between current and potential shareholders New shareholders often lack complete knowledge about a project's potential profits and risks, prompting them to demand shares at a price lower than the market value as compensation for future dividends This behavior increases financing costs, ultimately diminishing the value of investment projects with a positive Net Present Value (NPV) and discouraging investment decisions.
Many businesses struggle with limited financial resources, hindering their ability to invest in promising projects As a result, these firms often overlook growth opportunities, which ultimately diminishes their overall value.
Free cash flow theory, akin to underinvestment theory, examines how the availability of free cash flow within a company influences its investment decisions and overall business value When a business possesses ample cash flow, it can enhance its investment activities However, ineffective investments can lead to a decline in business value due to negative net present value (NPV) projects and associated agency costs Consequently, fluctuations in operating cash flow may result in a decrease in the company's value.
According to Jensen (1986), the free cash flow theory suggests that companies with excessive free cash flow may engage in suboptimal projects, exacerbating the conflict of interest between managers and shareholders, ultimately diminishing firm value Jensen defines free cash flow as the surplus cash remaining after funding all positive net present value projects, adjusted for the relevant cost of capital This dynamic creates information asymmetry between shareholders, who have ownership rights, and managers, who are responsible for daily operations (Berle & Means, 1932; Jensen & Meckling, 1976; Jensen, 1986; Shleifer & Vishny, 1997) Consequently, managers may prioritize their own interests over those of shareholders, leading to decisions that do not necessarily align with maximizing shareholder value.
DEFINITION anđ measurement
Substantial free cash flow often leads business managers to make suboptimal investment decisions, prioritizing personal interests over shareholder value This phenomenon, known as overinvestment, typically involves pursuing low-benefit or even negative NPV projects, ultimately harming shareholders and intensifying the conflict of interest between management and investors As a result, the overall value of the enterprise diminishes.
Richardson (2006) emphasized that managers can pursue multiple projects to meet their objectives and enhance benefits when a company enjoys positive or abundant free cash flow Conversely, companies facing negative free cash flow often struggle to do so, as they must seek external capital, which subjects them to market oversight (DeAngelo et al., 2004; Jensen).
In 1986, it was noted that utilizing free cash flow for project financing can lead to overinvestment and agency problems, as managers can easily avoid market oversight This situation necessitates higher agency costs, which are incurred from hiring external audit services to monitor and discipline managers who may misuse company funds on low-potential projects that diminish resources and overall value Jensen argued that a company with substantial free cash flow faces increased agency costs To mitigate this issue, limiting managerial power through debt financing is essential, as new equity issuance remains ineffective due to managers' influence over future cash flows via dividend decisions In contrast, loan contracts impose debt covenants and creditor oversight, making loan financing a more effective strategy to minimize agency costs and prevent overinvestment.
Excess free cash flow in firms can result in overinvestment in negative NPV projects, ultimately depleting resources and diminishing firm value Empirical studies by Vogt (1994), Richardson (2006), Park and Jang (2013), Badavar Nahandi and Taghizadeh Khanqah (2018), Ding et al (2019), Fu (2010), Liu and Bredin (2010), Titman et al (2004), and Yang (2005) provide consistent evidence supporting this theory.
Cash flow refers to the net amount of cash and cash equivalents entering and leaving a company It consists of cash inflows, which are the funds received, and cash outflows, which represent expenditures There are three main categories of cash flow: operating cash flow, investing cash flow, and financing cash flow.
Operating cash flow reflects the cash generated from a company's core operations, indicating its ability to sustain its activities Investing cash flow reveals the cash spent or earned from financial investments, while financing cash flow represents the net cash used to support the company's financial structure and capital needs Free cash flow is the cash remaining after capital expenditures and dividend payments, serving as an internal funding source for investment activities Cash flow volatility, as defined by Minton and Schrand (1999), refers to the fluctuations in a firm's operating cash flow over time, which can lead to reduced investments in capital expenditures, research and development, and advertising, a concept further explored by Qu (2020).
It presents the stability and predictability of cashflow Furthermore, it also depicts the future development space and the possibility of rising performance of a company
Changes in operating cash flows can arise from unpredictable factors such as economic fluctuations, political influences, customer behavior, and internal corporate policy adjustments Operating cash flow is crucial for making informed business investment decisions, as companies with sufficient cash flow can leverage internal funding rather than resorting to more expensive external sources like debt or new equity issuances The choice of funding source is influenced by the company's policies and operational circumstances While external funding can enhance liquidity during periods of limited cash flow, high costs associated with financial issuance may compel a business to rely on retained earnings or forgo new investment opportunities to sustain operations.
Various methods have been employed to assess cashflow volatility Minton and Schrand (1999) utilized the correlation coefficient with historical quarterly operating cashflow data from the six years prior to the base year, normalizing it by the absolute average of that data In contrast, Rountree et al (2008) measured cashflow volatility as the standard deviation of future quarterly operating cashflow fluctuations, while Shipe (2015) focused on the standard deviation of cash held by the enterprise Both Rountree et al (2008) and Mọkelọ (2012) argued that cashflow volatility serves as a reliable indicator of operating risk, as it is derived from operational cashflows rather than earnings, thereby minimizing the impact of discretionary accruals and potential manipulation or measurement errors on firm value.
Corporate investment plays a crucial role in shaping the perceptions of creditors, shareholders, and corporate managers regarding a business's performance and market position It encompasses expenses for maintaining current operations and pursuing new investment opportunities, including capital expenditures, research and development (R&D), acquisitions, and advertising costs (Minton & Schrand, 1999; Cohen, 2014; Panagiotidis & Printzis, 2021) Kallapur and Trombley (2001) distinguish between growth, which refers to an increase in size, and investment opportunities, which involve decisions to invest in value-enhancing projects Furthermore, corporate investment is linked to firm value and cash flow volatility (Njuguna et al., 2022); by investing in positive NPV projects, companies can boost revenues, generate additional cash, enhance market value, and reduce cash flow volatility.
Various methodologies have been employed to calculate corporate investment Minton and Schrand (1999) assessed investment costs by analyzing the total cost of capital, advertising expenses, and R&D expenses relative to total assets In contrast, Cohen (2014) and Richardson (2006) defined investment costs as the aggregate of capital costs, R&D expenditures, acquisition costs (excluding depreciation and amortization), and earnings from the sale of fixed assets Park and Jang (2013) adopted a similar approach to Cohen and Richardson but included depreciation and amortization costs, viewing them as essential for maintaining existing assets Additionally, their research, focusing on the service industry in Korea, substituted R&D costs with advertising expenses in the investment cost calculation.
In their 2017 study, investment costs were calculated by summing the change in capital stock and the depreciation of fixed assets for the year Conversely, Panagiotidis and Printzis (2021) determined investment costs by considering the net value of fixed assets along with the depreciation for that year.
According to Jensen and Meckling (1976), a firm functions as a tool designed to optimize inputs and outputs for maximum profit, specifically its present value Corporate managers are tasked with enhancing company value to ultimately optimize shareholder returns Thus, corporate value serves as a reflection of managerial capability.
The long-term development potential of an enterprise is influenced by various factors, including cash flow fluctuations, corporate investment, tangibility, profitability, and growth opportunities Firms with higher tangibility, or more tangible assets, can enhance revenue generation and secure debt at lower costs due to their ability to provide collateral (Chi & Su, 2017; Kondongo et al., 2014) Profitability serves as a crucial driver for increasing enterprise value, attracting both new and existing investors to further invest in the company, thereby boosting its overall worth (Rountree et al., 2008; Dang et al., 2021) Additionally, investing in growth opportunities allows companies to enhance both shareholder wealth and overall company value.
The Tobin Q index is widely used to appraise firm value, as it allows for comparisons between companies without the need for standardization or risk adjustments, calculated by the ratio of market to book value of assets (Chi & Su, 2017; Rountree et al., 2008; Mọkelọ, 2012; Kodongo et al., 2014) Additionally, firm value can be assessed through market value and stock returns (Gworo, 2019; Cai & Zhang, 2006) However, while these metrics can indicate growth potential, they are often unpredictable and influenced by external factors beyond managerial control (Bacidore et al., 1997) Moreover, stock prices are vulnerable to manipulation due to asymmetric information between managers and investors (Hax, 2003), making them an ineffective measure for evaluating enterprise value.
THE PREVIOUS RESEARCHES
Numerous empirical studies have explored the interrelationship between cash flow volatility, corporate investment, and firm value, yet their findings often lack consistency and sometimes conflict For clarity, previous research can be categorized into four distinct groups: the first examines the relationship between cash flow volatility and corporate investment; the second focuses on the link between cash flow volatility and firm value; the third investigates the relationship between corporate investment and firm value; and the fourth analyzes the mediating role of corporate investment in the connection between cash flow volatility and firm value A summary of these studies is presented in Table 2.1.
Table 2.1 Previous researches related to the interrelationships between Cashflow Volatility, Corporate Investments and Firm Value
Significant Source Data Span Empirical approach Findings
Relationship: The effect of cashflow volatility on firm value
The sample contains 3,501 firm-year observations for non-financial institutions from
Regression analysis; Cross- sectional, rolling window regressions
The findings investigated that cashflow volatility has an adverse impact on future firm value as it affects negatively to the future operating performance
A sample consists of 3,528 firms in 1987, 4,014 firms in
1992, and 4,449 firms in 1997 for a total of 11,991 observations from Compustat databases from 1987-2002
Univariate tests, Multivariate tests, Annual regressions, robustness tests
The empirical finding evidenced that cash-flow volatility has a negative effect on firm value and with a 1% increase in cash-flow volatility, resulting in approximately a 0.15% decrease in firm value
A data set of 778 European companies in total 2,211 firm year observations from 2000 to 2010
Pearson correlations, Univariate tests and Multivariate OLS regressions
The findings showed that both cash flow and earnings volatility seem to have a negative effect on firm value
US publicly-traded life insurers using derivatives for the purpose of hedging in schedule DB of the NAIC filing in the period of 2002-
Fixed effects regression, Regression-based Hausman, ordinary least squares (OLS) with firm-level fixed effect
The results illustrated that both derivatives hedging and cashflow volatility are negatively related to firm value
The sample of publically traded firms extracted from Compustat (excluded firms from industries of utilities_SIC code between
4900 and 4949 and financials _SIC code between 6000 and
Fama-MacBeth Regressions, pair-wise correlations, probit regression
Smaller firms, high-tech firms, and younger firms all display a stronger relationship between cash volatility and firm value than their respective counterparts
A sample includes 10,714 unique firms with over 80,000 firm-year observations in the period 1991 - 2012
Fixed effects effect regressions, Two-way clustering, Cross- sectional tests, Controlling for autocorrelation
The result evidenced a directly positive association between cash flow volatility and firm value (Tobin Q) Also, younger and smaller firms have higher performance volatility and higher firm valuation
A sample of 30 companies that were consistently listed in the NSE handbook in the period of 2011-2015
Multiple linear regression model, cross-sectional correlation design
The findings demonstrated that earning volatility and payout ratio have a positive effect on the market value of listed firms
A panel data model with 270- firm-year observations from the Jordanian commercial banks and insurance companies listed on Amman Stock Exchange (ASE) from
Fixed Effect Model (FEM) with Driscoll-Kraay standard errors
The empirical results depicted that cashflow s from operating activities per share had a positive and significant relationship with shareholder value, as an important indicator for firm value
Relationship: The effect of cashflow volatility on corporate investment
897 firms in 1989 and 1287 firms in 1995 in a period from
OLS regression, cross - sectional regressions and robustness checks
The findings provided that there is a negative and significant association between volatility cashflow and investment
A sample comprises 5,308 firm-year observations from Chinese listed companies of Shanghai and Shenzhen stock markets between 2005 and
The study reveals a significant negative relationship between cash flow uncertainty and corporate investment Companies experiencing higher cash flow uncertainty tend to adopt a more cautious approach to R&D innovation, while those with lower cash flow uncertainty are more inclined to invest boldly in innovative research and development initiatives.
An unbalanced data set of 25,000 larger Greek firms from the Infobank Hellastat database in the period 2000-
2014 two-step GMM, Markov Chain
Monte Carlo (MCMC) optimization method, quantile regression
The findings showed that the impact of uncertainty on the investment ratio is negative and statistically significant Also, this negative impact of uncertainty is more profound for smaller firms
An unbalanced panel data of
468 nonfinancial firms listed at the Pakistan Stock Exchange (PSX) during the period 2000–2018
(Generalized) autoregressive conditional heteroscedastic, (G)ARCH models, Robust two- step system generalized method of moments (2SYS GMM)
The results provided robust and strong evidence of the detrimental influence of all three types of uncertainty on investment
A sample consists of an unbalanced panel of 4,028 firm-years pertaining to 402 manufacturing firms for the period from 1984 to 2003
The findings indicate that intrinsic uncertainty leads to an increase in a firm's investment, making the investment rate more responsive to fluctuations in cash flow Conversely, extrinsic uncertainty negatively impacts the incentive to invest, regardless of cash flow levels.
A sample of 48,132 firm-year observations in
COMPUSTAT of U.S domestic firms during the period 1980–2012
OLS estimate, Fixed effects model, standard errors clustered
The finding pointed out that firms with high cash holdings increase investment when increasing cashflow volatility, whereas firms with low cash holdings lower their investment expenditure
An unbalanced firm-level of
23 nonfinancial firms listed on Kenya's NSE during the period 2000–2016
Random and Fixed effects, Generalized Autoregressive Heteroscedasticity, GARCH (1,1) technique
The results showed that uncertainty uncertainty has a positive effect on investments, which is consistent with the strategic growth options theory of investment
A sample of 177 South African listed firms from 1995 to 2017
Pooled Ordinary Least Squares (OLS), panel fixed-effects, two- step difference dynamic panel GMM
The findings demonstrated that there is a positive association between firm-specific uncertainty and investment
Relationship: The effect of investment on firm value
Samples consist of universal Canadian firms, which was 52 firms in 1985-1987, 54 firms in 1986-1988, and 47 firms in 1985-1988 in a period from
Ordinary-least-squares regressions with year dummies
The empirical results show a positive, statistically significant relationship between R&D spending and market value
A sample of 329 Chinese nonfinancial listed firms on SHSE or SZSE during 2004–
The findings indicate a non-monotonic relationship between managerial self-confidence and firm value, suggesting that there exists an optimal level of managerial confidence Additionally, the study reveals that firm value positively and significantly influences investments, whereas investments have a positive yet insignificant effect on firm value.
A sample of 563 Chinese listed firms between 2005 and
GMM estimator, pooled ordinary least squares (OLS) and fixed- effects panel regressions
The findings showed that R&D investments have an inverted U- shaped relationship with firm value, which depicted that when R&D investments increase, firm value increases to a certain level and then decreases
A sample data of listed non- financial firms on Vietnam' Stock Exchange from 2010 -
The outcomes showed that firm size, the ratio of investment for fixed assets, dividend payout ratio positively affect the firm value, while financial leverage,
21 growth rate of revenue and market value to book value ratio negatively affect firm value
A sample of 20 nonfinancial firms (200 observations) of four Arabic emerging markets during the period 2010–2019
Panel data regression, Fixed effects model, and Robustness Test
The finding indicated that increasing investment in marketing has a positive effect on the firm value valuation model
A sample of all companies listed on the Ho Chi Minh Stock Exchange (HOSE) and Hanoi (HNX) in the 11-year period 2009 - 2019
Fixed effects model, System GMM
Overinvestment has a statistically significant negative effect on firm performance
Relationship: The mediating effect of corporate investment on the link between cashflow volatility and firm value
A data set of 36 nonfinancial companies listed in Kenya from 2002 to 2019
Random effects panel regression model robust for standard errors was applied to analyse the data
A four-step mediation analysis revealed that corporate investments mediate the relationship between cash flow volatility and firm value The findings indicate that during uncertain times, companies tend to reduce their spending on corporate investments, which negatively impacts their overall value.
2.3.1 Previous researches about the relationship between cashflow volatility and firm value
There have been many researches about the link between cashflow volatility and firm value, however, they provided inconsistent results
Some emprirical studies have found evidence of a negative impact of cashflow volatility on firm value, including the findings of Minton et al (2002), Rountree et al
Numerous studies have established a negative relationship between cashflow volatility and firm value Minton et al (2002) conducted regression analysis on 3,501 firm-year observations from 1983 to 1997, revealing that cashflow volatility adversely impacts future operating performance and firm value Rountree et al (2008) confirmed this finding, noting that a 1% increase in cashflow volatility leads to a 0.15% decrease in firm value, based on a sample of 11,991 observations from Compustat databases between 1987 and 2002 Mọkelọ (2012) further supported these conclusions through Pearson correlations and OLS regressions on 2,211 observations from 778 European companies, indicating that both cashflow and earnings volatility negatively affect firm value In 2017, Altuntas et al examined 55 US publicly-traded life insurers and found that derivatives hedging and cashflow volatility are also negatively correlated with firm value Collectively, these empirical studies underscore the hypothesis that cashflow volatility significantly diminishes firm value.
Numerous studies, including those by Shipe (2015), Chi and Su (2017), Gworo (2019), and Sawalqa (2021), indicate that cash flow fluctuations positively influence company value Utilizing Fama-MacBeth Regressions, pair-wise correlations, and probit regression on publicly traded firms from Compustat (excluding utilities and financial sectors) from 1992 to 2013, Shipe (2015) found that smaller, high-tech, and younger firms exhibit a stronger correlation between cash volatility and firm value compared to their larger and more established counterparts.
The strategic decisions made by managers play a crucial role in aligning cash holdings to an optimal, often mobile target, which enhances business value despite increasing cash flow fluctuations Chi and Su (2017) conducted a comprehensive study using fixed effects regressions and two-way clustering on a sample of 10,714 unique firms, analyzing over 80,000 firm-year observations from 1991 to 2012 Their findings revealed a direct positive correlation between cash flow volatility and firm value, as measured by Tobin's Q Additionally, the research indicated that younger and smaller firms experience greater performance volatility and subsequently achieve higher firm valuations.
A study conducted in 2019 revealed that earning volatility and payout ratio positively influence the market value of listed firms This research utilized a sample of 30 companies consistently listed in the NSE handbook from 2011 to 2015, employing a multiple linear regression model and cross-sectional correlation design Additionally, Sawalqa (2021) utilized a Fixed Effect Model with Driscoll-Kraay standard errors and panel data from 270 firm-year observations of Jordanian commercial banks and insurance companies listed on the Amman Stock Exchange from 2011 to 2019, finding a positive and significant relationship between cash flows from operating activities per share and shareholder value, highlighting its importance as an indicator of firm value.
2.3.2 Previous researches about the relationship between cashflow volatility and corporate investment
Research on the correlation between cash flow volatility and corporate investment has produced inconsistent results, mirroring the relationship between cash flow volatility and firm value found in previous studies.
Numerous studies have demonstrated that cash flow volatility adversely affects corporate investment Research by Minton & Shrand (1999) utilizing OLS regression and cross-sectional analyses of 897 firms in 1989 and 1,287 firms in 1995 revealed a significant negative correlation between cash flow volatility and investment activities This volatility can hinder firms' access to capital markets and increase associated costs Subsequent studies by Beladi et al (2021), Panagiotidis and Printzis (2021), and Rashid et al (2022) have corroborated these findings, indicating that businesses experiencing high cash flow volatility tend to reduce their investment efforts.
A study by Beladi et al (2021) analyzed 5,308 firm-year observations from Chinese listed companies in the Shanghai and Shenzhen stock markets between 2005 and 2016, utilizing 2SLS and GMM regression methods to explore the relationship between cash flow uncertainty and corporate investment The findings revealed that cash flow uncertainty has a significant negative impact on corporate investment, with firms facing higher cash flow uncertainty tending to invest more conservatively in R&D innovation In contrast, firms with lower cash flow uncertainty are more willing to invest boldly in R&D The authors suggest that this negative relationship arises as firms aim to mitigate potential future losses by reducing their investment levels.
A study conducted in 2021 revealed a statistically significant negative impact of uncertainty on investment ratios, particularly affecting smaller firms more severely The research utilized an unbalanced dataset of 25,000 larger Greek firms from the Infobank Hellastat database, covering the period from 2000 to 2014, and employed advanced methodologies such as two-step GMM and quantile regression Furthermore, Rashid et al (2022) assessed the relationship between uncertainty and investment using an unbalanced panel data of 468 nonfinancial firms listed on the Pakistan Stock Exchange from 2000 to 2018, applying (G)ARCH models and robust two-step system generalized method of moments (2SYS GMM) Their findings provided strong evidence of the adverse effects of various types of uncertainty on investment decisions.
On the other hand, the study of Baum et al (2010), Cohen (2014), Kimaiyo
(2017), and Chortareas et al (2021) provided the opposite results that cashflow volatility has positive impact on corporate investment Specifically, the study of Baum et al
A study conducted in 2010 analyzed an unbalanced panel of 4,028 firm-years from 402 manufacturing firms between 1984 and 2003 using a GARCH model The findings revealed that intrinsic uncertainty positively influences capital investment, making investment rates more sensitive to fluctuations in cash flow Conversely, extrinsic uncertainty negatively affects the incentive to invest, regardless of cash flow levels.
A study conducted in 2014 analyzed the relationship between cash holdings and investment behavior using an OLS estimate and fixed effects model on a sample of 48,132 firm-year observations from U.S domestic firms between 1980 and 2012 The findings revealed that firms with substantial cash reserves tend to increase their investments in response to rising cash flow volatility, while those with lower cash holdings reduce their investment expenditures Similarly, Kimaiyo (2017) employed Random and Fixed effects along with a GARCH (1,1) technique to examine 23 non-financial firms listed on Kenya's NSE from 2000 to 2016, concluding that uncertainty positively influences investment decisions.
HYPOTHESIS FOR MODELS
Investment activities are essential for companies, regardless of whether their business situation is favorable or unfavorable, as they influence both current and future viability This study aims to explore how investment activities mediate the relationship between cash flow fluctuations and firm value To uncover this mediation effect, the dissertation will be structured into four key steps.
The first step involves analyzing the relationship between cash flow volatility and firm value, with evidence suggesting a significant negative correlation High volatility in operating cash flow indicates that a business is exposed to various operational risks, which can deter investors and ultimately reduce the company's value (Njuguna et al., 2022; Minton et al., 2002; Rountree et al., 2008; Mọkelọ, 2012; Altuntas et al., 2017) Thus, the first hypothesis is established.
H1: Cashflow volatility has a significant and negative impact on firm value
The dissertation will explore the relationship between cash flow volatility and corporate investment, positing that cash flow volatility is significantly and negatively correlated with a firm's investment activities Research by Fazzari et al (1987) indicates that firms' investment expenditures are sensitive to cash flow fluctuations Stable operating cash flows enable firms to finance projects with positive net present value, enhancing firm value In contrast, unstable cash flows compel businesses to seek external capital or reduce investments to maintain liquidity (Minton & Shrand, 1999; Beladi et al., 2021; Panagiotidis & Printzis, 2021; Rashid et al., 2022; Njuguna et al., 2022) Thus, the second hypothesis is established.
H2: Cashflow volatility has a significant and negative impact on corporate investment
In the third step, the author explores the relationship between corporate investment and firm value, highlighting that corporate investment is anticipated to significantly enhance firm value This is because investment activities are crucial for sustaining current operations and fostering future growth By investing in projects with positive Net Present Value (NPV), companies can boost their revenues and create long-term value.
28 more cash, which increases their market value (Johnson and Pazderka, 1993; Ye &Yuan (2008); Kim et al., 2018; Dang et al., 2020; Mousa et al., 2021; Njuguna et al., 2022), so the 3 rd hypothesis is:
H3: Corporate investment has a significant and positive impact on firm value
In the final step, the dissertation examines how cash flow volatility and corporate investment impact firm value, highlighting the mediating role of corporate investment in this relationship It posits that corporate investment negatively correlates with cash flow volatility while positively influencing firm value As noted by Njuguna (2022), investing in high-value projects enables firms to generate increased revenue, thereby reducing cash flow volatility and enhancing firm value Conversely, elevated cash flow volatility prompts companies to cut back on investment expenditures, ultimately diminishing firm value Thus, a strong connection exists between corporate investment, cash flow volatility, and firm value, leading to the main hypothesis.
H4: The mediating effect of corporate investments on the link between cashflow volatility and the firm value of non-financial firms listed on HOSE is significant
RESEARCH METHODOLOGIES
SAMPLE, DATA AND VARIABLES
The dissertation analyzes a sample of 300 Vietnamese non-financial firms listed on the HOSE between 2013 and 2022, representing 77.92% of the companies and approximately 70% of the total market capitalization of non-financial firms on the exchange.
The research utilizes secondary data from 300 non-financial firms, sourced from the Fiinpro X database and individual financial statements when necessary By excluding financial firms, the study analyzes a total of 1,500 firm-year observations spanning from 2018 to 2022, although the exact number of observations may differ depending on the specific model used.
This study analyzes the annually audited consolidated financial statements and quarterly consolidated financial statements of Vietnamese companies, which align with the annual and quarterly calendar.
This study examines six key variables: cashflow volatility (CFV), corporate investment (INV), firm value (FV), profitability (Profit), growth opportunity (MBVE), and tangibility (Tang) Among these, CFV, INV, and FV serve as the primary research variables, while Profit, MBVE, and Tang act as control variables Previous research indicates that Profit, MBVE, and Tang significantly influence company value, suggesting that their exclusion could skew the regression model's results.
First, cashflow volatility (CFV) is an independent variable in the research model
In the realm of measuring cashflow volatility, various methodologies exist, as highlighted in Chapter 2 Rountree et al (2008) defined cashflow volatility through the standard deviation of future quarterly operating cashflow fluctuations, while Shipe (2015) focused on the standard deviation of cash held by the enterprise This study will adopt the approach of Minton and Schrand (1999) and Njuguna et al (2022), with necessary adjustments to fit the current research context, specifically calculating cashflow variation based on these established frameworks.
The analysis examines the coefficient of variation in a firm's quarterly operating cash flow over the five years leading up to each sample year from 2018 to 2022 Each firm-year observation's coefficient of variation is adjusted relative to the median of all firms within the same industry for that year This method is chosen due to its accessibility and relevance, as operating cash flow is crucial for businesses to manage various expenditures (Charitou & Ketz, 1991) Data was primarily sourced from the Fiin proX database, with additional information gathered from the companies' quarterly consolidated financial statements to address any missing data.
Corporate investment (INV) serves as the mediating variable in this study, defined as the total of capital expenditures, advertising, R&D spending, depreciation, and amortization expenses, less the sale of property, plant, and equipment, divided by total assets This calculation method, based on research by Park and Jang (2013) and Njuguna et al (2022), is utilized due to the availability of data and its comprehensive coverage of investment costs necessary for enterprises to sustain and enhance their operations.
Investment costs encompass capital expenses, research and development costs, acquisition costs (excluding depreciation and amortization), and earnings from the sale of fixed assets In Vietnam, however, gathering data on acquisition costs poses challenges due to the market's weak efficiency.
2017), so the measurement becomes inappropriate The calculated figures were taken from the Fiinpro X database and the company's audited consolidated financial statements
Firm value (FV) serves as a critical indicator of a company's performance and management effectiveness, making it the dependent variable in this study This concept aligns with prior research conducted by Chi and Su (2017) and Rountree et al., highlighting the importance of understanding how management capabilities influence overall firm valuation.
The dissertation employs the Tobin Q ratio, calculated as the sum of market value of equity and book value of debt divided by total assets, to measure firm value (FV) (Mọkelọ, 2012; Kodongo et al., 2014; Njuguna et al., 2022) While firm value can also be gauged through market value or stock returns (Gworo, 2019; Cai & Zhang, 2006), these metrics are often unpredictable and influenced by external factors beyond managerial control (Bacidore et al., 1997) Additionally, stock prices are vulnerable to manipulation due to asymmetric information between managers and investors, rendering them an ineffective measure for evaluating enterprise value (Hax, 2003).
The author analyzed enterprise value by controlling three key variables: tangibility (Tang), defined as the ratio of fixed assets to total assets; profitability (Profit), calculated as the ratio of earnings before interest and tax to total assets; and growth opportunity (MBVE), measured by the ratio of market value of equity to total shareholder's equity.
According to Su (2017) and Kondongo et al (2014), firms with a higher proportion of tangible assets enhance their revenue generation capabilities and secure debt at reduced costs due to their ability to provide collateral, indicating that tangibility (Tang) positively influences firm value (FV) Profitability (Profit) also plays a crucial role in boosting enterprise value, attracting both new investors and existing shareholders to invest further in the company, thereby enhancing its value (Rountree et al., 2008; Dang et al., 2021) Additionally, growth opportunities (MBVE) are considered a control variable in this study, as they are expected to positively correlate with firm value (FV); investments in profitable opportunities can increase both shareholder wealth and overall company value The calculation of growth opportunities (MBVE) follows the methodology of Njuguna et al (2022), reflecting investor perceptions of the company's growth potential.
Removing essential variables can lead to model failure and inaccurate results Detailed definitions and resources for these variables are provided in Appendices 1 and 2 The interrelationships among the variables discussed in the dissertation are illustrated in the diagram below.
Figure 3.1 The interrelationships between the variables used in the research
RESEARCH MODELS
This dissertation employs the four-step mediation model of Baron and Kenny (1986) to investigate whether corporate investment (INV) serves as a mediating factor between cash flow volatility (CFV) and firm value (FV) The research process systematically addresses four key steps, corresponding to the research questions and hypotheses, to elucidate the relationship between these variables.
Step 1: Examining the link between cashflow volatility (CFV) and firm value (FV)
Step 2: Investigate the link between cashflow volatility (CFV) and corporate investment (INV)
Step 3: Examining the link between corporate investment (INV) and firm value (FV)
Step 4: Investigate the influence of cashflow volatility (CFV) - independent variable and corporate investment (INV) - mediator variable on firm value (FV) - response variable
To enhance the modeling of relationships between variables in simple linear regression and improve result reliability, all variables in the dissertation were transformed into natural logs This transformation, based on the study by Njuguna et al (2022), aimed to convert raw skewed data into a more normally distributed form, thereby stabilizing variance and reducing heteroscedasticity.
3.2.1 Step 1 - Examining the link between cashflow volatility (CFV) and firm value (FV)
Baron and Kenny (1986) established that a mediating relationship requires a connection between independent and dependent variables In this study, hypothesis H1 posits a significant correlation between cashflow volatility (CFV) as the independent variable and firm value (FV) as the dependent variable To analyze this relationship, a panel regression model is employed, incorporating the natural logs of CFV and FV along with control variables such as profitability (Profit), growth opportunity (MBVE), and tangibility (Tang) The regression model is articulated as follows: lnFV i,t = 𝛽0 + 𝛽1 lnCFV i,t + 𝛽2 lnProfit i,t + 𝛽3 lnMBVE i,t + 𝛽4 lnTang i,t + 𝜀i,t (Model 1).
The article analyzes the relationship between firm value, represented by the natural log of the Tobin Q ratio (lnFV i,t), and cash flow volatility, indicated by the natural log of the adjusted coefficient of variation (lnCFV i,t) Additionally, it incorporates control variables such as profitability (lnProfit i,t), growth opportunity (lnMBVE i,t), and tangibility (lnTang i,t), all derived from data collected in year t As discussed in Chapter 2, the correlation between the main independent variable and the dependent variable is explored to understand their interdependencies.
The expected relationship between control and dependent variables indicates a positive correlation, while cash flow fluctuations are anticipated to have a negative effect on corporate value, as supported by studies from Rountree et al (2008), Mọkelọ (2012), Minton et al (2002), and Njuguna et al (2022) Minton et al (2002) highlighted that cash flow fluctuations can lead to underinvestment issues, adversely affecting corporate value However, contrasting findings from Shipe (2015), Gworo (2019), and Sawalqa (2021) suggest that cash flow fluctuations may actually enhance company value This indicates that the model's outcomes can vary based on the research data, provided there is a significant relationship between the variables For detailed definitions and the predicted relationships of these variables, please refer to Appendix 3.
3.2.2 Step 2 - Investigate the link between cashflow volatility (CFV) and corporate investment (INV)
To establish an intermediating relationship, it is essential that the mediating variable is significantly affected by the independent variable Specifically, cashflow volatility (CFV) must exert a notable influence on corporate investment (INV), as outlined in hypothesis H2 Consequently, this study conducts a regression analysis focused solely on the relationship between corporate investment (INV) and cashflow volatility (CFV), represented by Model 2: lnINV i,t = 𝛽0 + 𝛽1 lnCFV i,t + 𝜀i,t.
In this study, lnINV i,t represents corporate investment, calculated as the natural logarithm of the total capital expenditures, advertising, R&D spending, depreciation, and amortization, minus property, plant, and equipment sales, divided by total assets, serving as the mediating variable while also acting as the dependent variable lnCFV i,t denotes cashflow volatility, an independent variable measured by the natural logarithm of the adjusted coefficient of variation for each firm-year observation Based on prior studies by Minton and Shrand (1999), Beladi et al (2021), Rashid et al (2022), and Njuguna et al (2022), a negative relationship between these two variables is anticipated, although previous findings regarding cashflow volatility and firm value have shown inconsistency Despite this, the primary objective of step 2 is to ascertain the significant relationship between lnINV and lnCFV, making the sign correlation less critical Definitions and predicted relationships of the variables are detailed in Appendix 4.
3.2.3 Step 3 - Examining the link between corporate investment (INV) and firm value (FV)
To establish a mediating relationship, it is essential to create a significant connection between the mediating variable, corporate investment (INV), and the dependent variable, firm value (FV) As outlined by Baron and Kenny (1986), hypothesis H3 posits that corporate investment must have a substantial effect on firm value This relationship is represented in the regression equation of Model 3: lnFV i,t = 𝛽0 + 𝛽1 lnINV i,t + 𝛽2 lnProfit i,t + 𝛽3 lnMBVE i,t + 𝛽4 lnTang i,t + 𝜀i,t.
The article discusses the relationship between corporate investment (INV) and firm value (FV), utilizing the natural logarithm of the Tobin Q ratio to represent FV and the natural log of various expenditures (capital, advertising, R&D, depreciation, and amortization) adjusted for asset sales to define INV Control variables include profitability (Profit), growth opportunity (MBVE), and tangibility (Tang), all measured in natural logs Data for these variables were collected in year t, and based on previous studies, it is anticipated that increased corporate investment will positively and significantly impact firm value, with control variables also expected to positively correlate with the explanatory variable Further definitions and predicted relationships of these variables are provided in Appendix 5.
3.2.4 Step 4 - Investigate the influence of cashflow volatility (CFV) - independent variable and corporate investment (INV) - mediator variable on firm value (FV) - response variable
This crucial step of the dissertation aims to evaluate hypothesis H4, focusing on the effects of explanatory and mediating variables on the response variable Unlike previous steps, this analysis requires a significant interrelationship among variables, with the main independent variable, cashflow volatility (CFV), demonstrating lower statistical significance than in step 1 Additionally, the absolute value of the regression coefficient (𝛽) for CFV must be smaller in this step compared to step 1 to confirm the mediating relationship, as outlined by Baron and Kenny (1986) The regression equation is detailed in Model 4.
35 lnFV i,t = 𝛽0 + 𝛽1 lnCFV i,t + 𝛽2 lnINV i,t + 𝛽3 lnProfit i,t + 𝛽4 lnMBVE i,t + 𝛽5 lnTang i,t + 𝜀i,t
The article discusses the relationships between various financial metrics for company i in year t, including firm value (FV), cashflow volatility (CFV), corporate investment (INV), profitability (Profit), growth opportunity (MBVE), and tangibility (Tang), represented as natural logs (ln) In this model, lnFV serves as the dependent variable, while lnCFV is the primary independent variable, and lnINV acts as a mediator The control variables include lnProfit, lnMBVE, and lnTang The interrelationships among these variables, as outlined in Chapter 2, remain consistent with previous models It is anticipated that the coefficient for cashflow volatility (CFV) will be less significant than in the initial research model Detailed definitions of these variables and their expected relationships are provided in Appendix 6.
RESEARCH METHODS
3.3.1 Descriptive statistics and Pearson correlation analysis
Descriptive statistics play a crucial role in various studies by providing concise coefficients that represent a population or its sample They are categorized into two main types: Measures of Center, which include mean, median, and mode to highlight the central tendencies of a data set, and Measures of Variation, which encompass standard deviation, variance, minimum and maximum values, as well as range, quartiles, deciles, percentiles, and the five-number summary These statistical measures effectively summarize and illustrate the key characteristics of data sets (Illowsky et al., 2018).
Pearson correlation analysis, commonly used to assess linear relationships, provides an initial understanding of the statistical connections between variables in research, following the methodology of Njuguna et al (2022) By utilizing descriptive statistics alongside Pearson correlation analysis, researchers gain a comprehensive overview of the variables, facilitating a more efficient and streamlined process for analysis and evaluation.
The study employs panel regression to examine the effects of various variables, building on previous research while highlighting the model's significant advantages Panel data analysis offers a robust framework for understanding complex relationships within the data.
Panel data can effectively manage unobservable, non-computable, or unavailable variables that correlate with independent or predictive variables, addressing the challenge of missing data due to heterogeneity (Baltagi, 2008).
The author employed a random effects regression model (REM) and a fixed effects model (FEM), along with the System-Generalized Method of Moments (system-GMM), following the four-step mediation model established by Baron and Kenny (1986) The choice of regression methodology was adaptable based on the specific circumstances of each model Notably, the system-GMM was utilized as a final recourse when earlier regression models failed to yield stable, consistent, or reliable results A typical panel regression model is represented by a specific regression equation.
In the given model, y represents the dependent variable, while x serves as the independent variable The correlation coefficient is denoted by β, and α acts as the intercept factor Additionally, i and t indicate the indices for each individual and time, respectively, with ε i,t representing the model's error.
The random effects model (REM) is widely utilized in research, particularly in studies where the characteristics of observation units do not influence the independent variable The standard regression equation for REM is typically presented as follows:
In this article, y represents the dependent variable, while x denotes the independent variable The indices i and t correspond to individual and time, respectively The intercept for each individual is denoted by αi, and β indicates the correlation coefficient Additionally, εi,t reflects the error related to the individual characteristics over time, and 𝑢i signifies the random error for each object.
The Random Effects Model (REM) assumes that model errors do not influence the independent or explanatory variables As noted by Torres-Reyna (2007), this model is particularly useful when variations among research entities affect the dependent variable, allowing for broader generalizations beyond the selected sample Additionally, numerous studies highlight the REM model's cost-effectiveness compared to the Fixed Effects Model (FEM) in terms of required parameters Furthermore, the REM model is appropriate when the intercept of each cross-sectional unit does not impact the independent variables, aligning with the model's foundational assumptions regarding individual error.
In the sample of 37, there is no correlation with the independent or explanatory variables Factors that remain constant over time can effectively serve as explanatory variables in the model (Borenstein et al., 2010).
The Fixed Effects Model (FEM) is widely utilized in panel data analysis, particularly when researchers anticipate that individual characteristics influence both the independent and dependent variables, unlike the Random Effects Model (REM) FEM effectively controls for unique, time-invariant traits of the subjects under study, provided there is no interaction between these traits This model is specifically designed to eliminate the effects of time-invariant characteristics, enabling a clearer assessment of how independent variables impact dependent variables The regression equation for the one-way Fixed Effects Model is as follows:
In this model, y represents the dependent or outcome variable, while x serves as the independent or predictor variable The indices i and t denote each individual and time, respectively The intercept for each individual is denoted by αi, and β represents the correlation coefficient Additionally, εi,t accounts for the error linked to the individual characteristics over time.
In econometric modeling, the distinction between the Fixed Effects Model (FEM) and the Random Effects Model (REM) lies in their treatment of random errors The FEM assumes that the error term, ε i,t, represents a non-random change specific to each object, whereas the REM posits that ε i,t is a random change This fundamental difference influences the choice of model based on the nature of the data and the underlying assumptions about the error structure.
The Fixed Effects Model (FEM) regression is utilized by researchers to analyze the effects of time-varying variables on observed entities According to Torres-Reyna (2007), this model helps to examine the relationship between independent (predictor) variables and dependent (outcome) variables, considering how individual characteristics influence these predictors The FEM model is based on two key assumptions: first, that the errors associated with observed individuals affect the predictor or dependent variable; and second, that the invariant characteristics of each individual are not correlated with the unique characteristics of other individuals in the observation group (Borenstein et al.).
2010) In other words, the error of the observations (ε i,t) and the model's intercept (αi) should not affect each other (Torres-Reyna, 2007)
System-GMM is a popular choice in prior research examining the individual impacts of cash flow fluctuations (CFV), corporate investment, and firm value (FV), due to its effectiveness in addressing endogeneity issues A notable example is the 2021 study by Beladi et al., which explored the connection between CFV and investment patterns among Chinese firms.
EMPERICAL RESULTS
DESCRIPTIVE STATISTICS
The author aims to provide a comprehensive overview of data usage by employing a combination of descriptive statistics, specifically Measures of Center and Measures of Variation The study presents key statistical parameters, including the number of observations, mean value, standard error, minimum value, and maximum value for all relevant variables To facilitate tracking and comparison, these parameters are organized in a table, accompanied by detailed explanations of the variables.
Table 4.1 Overall Summary Statistics of Study Variables
Variables Obs Mean Std dev Min Max
This article analyzes firm characteristics across 1,500 observations, including 300 listed firms on HOSE from 2018 to 2022 Key variables examined include Firm Value (FV), calculated using Tobin's Q, and Cashflow Volatility (CFV), measured by the adjusted coefficient of variation in quarterly operating cashflow over the preceding five years Corporate Investment (INV) is assessed as the net sum of capital expenditures, advertising, R&D, and depreciation, scaled by total assets Profitability (Profit) is defined as the ratio of earnings before interest and tax to total assets, while Tangibility (Tang) represents the proportion of fixed assets to total assets Finally, Growth Opportunity (MBVE) is computed as the ratio of market value of equity to total shareholder's equity.
Firm value (FV) is assessed using the Tobin Q index, which measures the ratio of market value to book value of a company's assets, as demonstrated in studies by Rountree et al (2008), Mọkelọ (2012), Kodongo et al (2014), and Chi and Su (2017) A Tobin Q greater than one indicates that a firm is overvalued, suggesting that investors are willing to pay more for its assets than their current worth, whereas a Tobin Q below one signifies undervaluation, reflecting investor uncertainty During the analysis period, the average Q-index for listed non-financial companies was 1.207, with a maximum value of 13.940, indicating generally positive market evaluations despite some undervalued firms The standard error of 0.906 suggests that most companies are highly assessed, with minimal variance in market valuations This finding aligns with Dang et al (2021), which reported a mean Tobin's Q of 1.129 and a standard error of 0.668 for listed companies in Vietnam from 2006 to 2017.
Cashflow volatility (CFV) indicates the fluctuations in a company's operating cashflow, serving as a measure of operational risk and the capacity to generate revenue or cover expenses The mean and standard error of CFV for non-financial companies listed on the HOSE are 3,464 and 14,667, respectively, highlighting significant disparities in performance This data suggests that cashflow volatility is not concentrated around the average, with maximum CFV reaching 309,650 and a minimum of 0.000 in certain years Such drastic fluctuations may reflect uncertainties during the analysis period, notably influenced by the Covid pandemic's impact on non-financial businesses in Vietnam and the broader Vietnamese economy.
Corporate investment (INV) refers to the expenses an enterprise allocates for maintaining and developing operations, expressed as a ratio of total assets According to Table 4.1, the average non-financial company on the HOSE invests approximately 4.1% of its asset value The standard deviation of the investment-to-asset ratio is 6.5%, indicating significant variability compared to the mean This suggests that the investment rate among non-financial enterprises listed on the HOSE is relatively low, highlighting a notable fluctuation in their investment behaviors.
The investment ratios of companies vary significantly, with a maximum level reaching 97% and a minimum of 0% in relation to their assets This diversity in investment levels among the analyzed firms highlights the distinctions between capital-intensive and non-capital-intensive companies, offering valuable insights into their financial dynamics.
The profitability ratio, calculated as EBIT divided by total assets, highlights a company's ability to generate income and manage operating costs On average, non-financial companies listed on the HOSE generate VND 7.4 in operating revenue for every VND 100 of assets, indicating inefficiencies in profitability since this measure excludes interest and taxes, which are critical to assessing actual profit-generating capability Njuguna et al (2022) note that a low profitability ratio reflects poor performance, as operating revenue fails to account for fixed financing costs and taxes that are unavoidable for businesses The variability in profitability, with a standard error of 8.8%, underscores the differences in revenue-generating capabilities among companies, where the highest profitability ratio is 78%—indicating VND 78 in revenue from VND 100 of assets—while the lowest is -36%, signifying a loss of VND 36 per VND 100 of assets.
Tangibility (Tang) plays a crucial role in estimating a business's debt capacity, as fixed assets can serve as collateral, thereby lowering lending costs (Njuguna et al., 2022) This metric also aids in evaluating a company's ability to mitigate distress costs during financial difficulties According to Table 4.1, the average tangibility value is 24.7%, indicating that out of every VND 100 in assets, VND 24.7 consists of fixed assets available for collateral However, this average does not imply that firms possess minimal fixed assets, as evidenced by a standard deviation of 0.214, highlighting significant variability Furthermore, the distribution of fixed assets varies across different industries, with capital-intensive sectors like manufacturing, construction, and energy exhibiting distinct asset structures compared to non-capital-intensive sectors such as services and commercial services.
A higher MBVE ratio, particularly one exceeding 1, indicates that investors hold a strong belief in the company's future growth potential, suggesting that the business is highly valued or possibly overvalued Conversely, a negative growth opportunity points to inefficiencies within the company.
The analysis reveals that the market often perceives companies as either undervalued or overvalued, despite stable operations The average stock price during the study period is 1,396 times higher than the book value, indicating a general market overvaluation Notably, there is a wide disparity in valuations, with a minimum of -7,240 times and a maximum of 61,670 times, resulting in a standard deviation of 2,168 times This variation highlights the differing perceptions of stocks based on individual company performance and potential.
PEARSON CORRELATION ANALYSIS - CORRELATION MATRIX
The author employed a correlation matrix to assess the strength and linear relationships among the variables, utilizing four distinct matrices aligned with the research model at each stage of Baron and Kenny's (1986) four-step mediation model These matrices are detailed in Appendices 7, 8, 9, and 10.
In Appendix 7, Model 1, step 1 utilizes variables such as firm value (FV), cashflow volatility (CFV), profitability (Profit), growth opportunity (MBVE), and tangibility (Tang), represented as lnFVi,t, lnCFVi,t, lnProfit i,t, lnMBVE i,t, and lnTang i,t The analysis reveals that most correlations between firm value and the other variables are positive, with the exception of the negative relationship between firm value and cashflow volatility Conversely, cashflow volatility exhibits negative associations with all other variables Additionally, the control variables of profitability, growth opportunity, and tangibility demonstrate positive interrelationships.
The correlations among the independent variables exhibit a wide range, with absolute values fluctuating between 0.15 and 0.92 The strongest correlation is observed between lnFVi,t and lnMBVE i,t, while the weakest correlation occurs between lnMBVE i,t and lnTang i,t.
Appendix 8 presents a correlation matrix illustrating the relationship between corporate investment (lnINV i,t) and cashflow volatility (lnCFV i,t) in Model 2, step 2 The analysis reveals a negative correlation of -0.1085 between these two variables, aligning with the initial expectation of an inverse relationship.
Appendix 9 shows the correlation coefficient between all the variables including firm value (FV) - lnFVi,t, corporate investment (INV) - lnINV i,t , profitability (Profit) - lnProfit i,t , growth opportunity (MBVE) - lnMBVE i,t , tangibility (Tang) - lnTang i,t in Model 3 In general, all correlations between the variables are positive and the correlation coefficients range from 0.13 - 0.92 The positive linear relationship between firm value (FV) - lnFVi,t and corporate investment (INV) - lnINV i,t is at 0.1547
The correlation matrix presented in Appendix 10 reveals significant relationships among various financial variables, including firm value (FV), cashflow volatility (CFV), corporate investment (INV), profitability (Profit), growth opportunity (MBVE), and tangibility (Tang) Notably, the correlation between firm value (FV) and growth opportunity (MBVE) is particularly strong, with a coefficient of 0.9197, indicating a positive association Conversely, the relationship between firm value (FV) and cashflow volatility (CFV) is negative, with a coefficient of -0.2626, suggesting that increased cashflow volatility may adversely affect firm value Similar positive correlations are observed with corporate investment (INV) and profitability (Profit).
The correlation coefficients between various financial metrics reveal important insights into corporate investment and firm value Specifically, the correlation between corporate investment (INV) and firm value (FV) is 0.1547, while the relationship between profitability (Profit) and firm value (FV) is marked by a coefficient of 0.1339 Additionally, cash flow volatility (CFV) shows a correlation of 0.1618 with firm value, while growth opportunity (MBVE) and tangibility (Tang) exhibit coefficients of 0.2460 and -0.1085, respectively These correlations highlight the interconnectedness of profitability, investment, and firm valuation within the corporate finance landscape.
The analysis reveals significant correlations among various financial metrics, with growth opportunity (MBVE) showing a strong positive correlation of 0.9197 with firm value (FV) and a negative correlation of -0.2596 with cashflow volatility (CFV) Corporate investment (INV) exhibits a negative correlation with other variables, indicating an inverse relationship with both cashflow volatility and firm value, suggesting that increased cashflow volatility may adversely affect corporate investment and firm value during the research period Additionally, tangibility (Tang) displays weak positive correlations with corporate investment and firm value, further emphasizing the complex interplay between these financial indicators.
The correlation coefficients among the variables in the models are all non-zero, indicating significant statistical relationships essential for applying Baron and Kenny's (1986) research model to explore the mediating effects of corporate investment (INV) on the relationship between cash flow volatility (CFV) and firm value (FV) Most relationships among the variables are positive, although the correlation between cash flow volatility (CFV) and other variables is negative Consequently, the signs of the coefficients in the results align with the expected relationships among the research variables in the regression models.
REGRESSION ANALYSIS BY FEM AND REM
In this section, the author presents an overview of the results of each approach, namely FEM and REM, in the implementation of four models corresponding to 4 steps
45 of the main research model Specifically, this section will divide into two sub-sections corresponding to the two approaches, the FEM and REM regression models
4.3.1 The regression results of 4-step mediation detecting models by FEM
The author employs four models—Model 1, Model 2, Model 3, and Model 4—to illustrate the mediation effect of corporate investment (INV), following the methodology established by Baron and Kenny (1986) The regression results derived from the Fixed Effects Model (FEM) method are presented in the subsequent table.
Table 4.2 Regression results of 4 models by FEM
Step 1-Model 1 Step 2-Model 2 Step 3-Model 3 Step 4-Model 4 VARIABLES Y = lnFV Y = lnINV Y = lnFV Y = lnFV lnCFV -0.002 -0.010 -0.004
This table illustrates the relationship between independent and dependent variables while incorporating control variables, excluding Model 2, using Fixed Effects Model (FEM) The dependent variables across the models are firm value (FV) represented as lnFV for Models 1, 2, 3, and 4, and corporate investment (INV) represented as lnINV for Model 3 The primary independent variables include cashflow volatility (CFV) denoted as lnCFV for Models 1 and 2, and corporate investment (INV) for Model 3, while Model 4 features both cashflow volatility (CFV) and corporate investment (INV) Control variables include profitability (Profit) as lnProfit, growth opportunity (MBVE) as lnMBVE, and tangibility (Tang) as lnTang The number of observations for the period from 2018 to 2022 is 1,199 for Model 1, 978 for Model 2, 934 for Model 3, and 893 for Model 4, with variations due to the application of the natural logarithm Standard errors are clustered by firm and presented in parentheses, with significance levels indicated at 10%, 5%, and 1% by *, **, and ***, respectively.
The analysis of Model 1 reveals a negative correlation between cashflow volatility (CFV) and firm value (FV), while profitability, growth opportunity, and tangibility exhibit positive correlations with firm value Notably, growth opportunity shows a significant positive impact on company value with a coefficient of 0.499 at a 1% significance level In Model 2, cashflow volatility has a negligibly negative correlation with corporate investment Model 3 indicates that corporate investment has a non-significant negative correlation with firm value, while profitability, growth opportunity, and tangibility positively influence business value with significant coefficients of 0.020, 0.508, and 0.022, respectively Finally, Model 4's results indicate that both cashflow volatility and corporate investment have negative correlations with firm value, but these are not statistically significant, whereas all control variables maintain positive and statistically significant correlations at the 10% level or higher.
In conclusion, the findings from the FEM analysis largely align with the anticipated relationships among the research variables, except for the unexpected negative correlation between corporate investment (INV) and firm value (FV), which lacks statistical significance Conversely, the control variables demonstrated significant correlations with firm value (FV), exhibiting relatively high levels of significance.
4.3.2 The regression results of 4-step mediation detecting models by REM
With the same objectives as above, the author continues to apply the REM method to regress Model 1, 2, 3, and 4 Regression results are shown in the below table:
Table 4.3 Regression results of 4 models by REM Step 1-Model 1 Step 2-Model 2 Step 3-Model 3 Step 4-Model 4
VARIABLES Y = lnFV Y = lnINV Y = lnFV Y = lnFV lnCFV -0.003 -0.056** -0.005
This table illustrates the relationship between independent and dependent variables while incorporating control variables, except for Model 2, using Random Effects Model (REM) The dependent variables across the models include firm value (FV) represented as lnFV for Models 1, 2, 3, and 4, and corporate investment (INV) shown as lnINV for Model 3 The primary independent variables consist of cashflow volatility (CFV) indicated as lnCFV for Models 1, 2, and 4, alongside corporate investment (INV) in Model 3 and Model 4 Control variables include profitability (Profit) as lnProfit, growth opportunity (MBVE) as lnMBVE, and tangibility (Tang) as lnTang The observation counts for the years 2018-2022 are 1,199 for Model 1, 978 for Model 2, 934 for Model 3, and 893 for Model 4, with variations attributed to the application of natural logarithms Standard errors are clustered by firm and presented in parentheses, with significance levels marked at 10%, 5%, and 1% as *, **, and ***, respectively.
First, for Model 1, the results of REM method show that cashflow volatility (CFV)
The analysis reveals that cash flow volatility (lnCFV) negatively correlates with firm value (lnFV), while profitability (lnProfit) and growth opportunity (lnMBVE) exhibit positive correlations with firm value, both significant at the 1% level with coefficients of 0.016 and 0.522, respectively This indicates that higher profitability and growth opportunities positively influence company value Additionally, in Model 2, cash flow volatility (lnCFV) is shown to have a statistically significant negative correlation with corporate investment (lnINV).
48 response variable in the model The correlation coefficient of cashflow volatility (CFV)
The analysis reveals that cashflow volatility (CFV) exhibits a negative influence on corporate investment (INV), with a coefficient of -0.056 that is significant at the 5% level In Model 3, corporate investment (INV) shows a negative but statistically insignificant correlation with firm value (FV) Conversely, the control variables—profitability (Profit), growth opportunity (MBVE), and tangibility (Tang)—demonstrate positive correlations with firm value, significant at the 1% and 5% levels, with coefficients of 0.029, 0.523, and 0.015, respectively This indicates that profitability, growth opportunities, and tangibility enhance business value over the five-year period Finally, Model 4's REM regression results confirm that both cashflow volatility (CFV) and corporate investment (INV) have negative correlations with firm value (FV), although these findings are not statistically significant In contrast, profitability, growth opportunity, and tangibility maintain positive correlations with firm value, significant at the 5% level or higher.
In summary, the regression results align with most expectations regarding the main research variables, except for the insignificant negative relationship between corporate investment (INV) – lnINV and firm value (FV) – lnFV Conversely, the control variables demonstrated a significant correlation with firm value (FV) – lnFV at a relatively high significance level These findings are consistent with those from the fixed effects model (FEM), with the exception of Model 2, which revealed a significant negative relationship between cash flow volatility (CFV) – lnCFV and corporate investment (INV) – lnINV.
TEST FOR CONSISTENCY AND UNBIASEDNESS
The results from the FEM and REM analyses generally lack statistical significance, with the exception of the REM regression for Model 2 in Step 2, which highlights a significant negative effect of cash flow volatility (CFV) on corporate investment (INV) To reach definitive conclusions and conduct further analysis, it is essential for the models to undergo statistical hypothesis testing to verify their consistency and unbiasedness Ensuring that all models are free from issues is crucial for the reliability and applicability of the findings in future research.
In Chapter 3, the author must decide between the Finite Element Method (FEM) and the Regression Equation Method (REM) to determine the most appropriate approach for each phase of the Baron and Kenny (1986) model for identifying mediated effects.
The Hausman test, introduced by Durbin in 1954, was utilized to analyze the significance of the main independent variables' coefficients in the dissertation This examination aims to identify the relationships between the variables, with the results detailed in the table below.
Table 4.4 Hausman test for choosing between FEM and REM
Chosen model FEM REM FEM FEM
Research variable at significance at No Yes (5%) No No
The Hausman test results indicate that the p-value for most models is below 0.05, with the exception of Model 2 in step 2 For Models 1, 3, and 4, the null hypothesis (H0), which posits that the difference between coefficients is not systematic, is rejected, suggesting that the Fixed Effects Model (FEM) is more appropriate for these models Conversely, Model 2 shows an insignificant p-value of 0.0831, leading to the acceptance of the null hypothesis and indicating that the Random Effects Model (REM) is better suited for this regression Additionally, the explanatory variable's significance level at 5% confirms a correlation between the variables in step 2 Consequently, the author recommends using the FEM for Models 1, 3, and 4, while opting for the REM for Model 2.
To enhance the consistency and reliability of the selected models for final conclusions, the research will next focus on identifying and addressing the issues faced by these models This dissertation has conducted several hypothesis tests, including heteroskedasticity, autocorrelation, and endogeneity tests, to evaluate model performance The findings from these tests are detailed in Table 4.5.
Table 4.5 Heteroskedasticity, Autocorrelation, Endogeneity Test
Heteroskedasticity Yes Yes Yes Yes
Auto-correlation Yes Yes Yes Yes
Endogeneity Yes No Yes Yes
Chosen model to fix System-GMM REM with clustered System-GMM System-GMM
The results in Table 4.5 indicate that most models exhibit Heteroskedasticity, Auto-correlation, and Endogeneity, with p-values below 0.05, rendering their outcomes unreliable Consequently, the study will implement correction methods to address these issues Specifically, the System-GMM approach, developed by Arellano and Bover (1995) and Blundell and Bond (1998), will be utilized for models affected by Endogeneity For Model 2, the Random-effects GLS regression method with clustered standard errors, as designed in Stata 17 software, will be applied to eliminate errors within the model.
REM WITH CLUSTERED STANDARD ERRORS, SYSTEM-GMM RESULTS AND
The selected models failed to address the issues identified in the research framework, leading to inconsistent and biased regression results Consequently, they are deemed unsuitable for further analysis and conclusions To rectify this, the author employed Random Effects Model (REM) with clustered standard errors for Model 2 and System Generalized Method of Moments (System-GMM) for the other models This section provides a detailed presentation and interpretation of the results obtained from REM with clustered standard errors and System-GMM.
4.5.1 System-GMM for Step 1 - Model 1
To validate the results of the System-GMM model, it is essential that they meet the specified conditions for diagnostic tests and persistence Key requirements must be fulfilled concurrently to ensure the reliability of the findings.
To ensure the reliability of the model, the Hansen test must yield a valid result, indicated by a p-value greater than 0.05, signifying insignificance This assessment is crucial for confirming that the results are unbiased However, it is important to note that a higher p-value does not necessarily imply a better outcome As Roodman (2009) cautioned, one should not take comfort in the Hansen test due to potential risks.
A p-value below 0.1 indicates a strong result, while values equal to or greater than 0.25 may signal potential issues Additionally, a p-value of 1.00 can compromise the reliability of the Hansen test results For optimal stability and reliability, the p-value should ideally fall between 0.1 and 0.25.
The System-GMM regression model yields reliable results when the first Arellano-Bond test for residual autocorrelation is significant, indicating the presence of autocorrelation Conversely, if the null hypothesis of the second Arellano-Bond test is accepted, it suggests no autocorrelation This assessment is based on comparing the test's p-value to the 0.05 threshold; a p-value greater than 0.05 indicates the absence of autocorrelation, while a p-value below 0.05 suggests its presence (Nguyen Doan Man, 2016).
Roodman (2009) emphasized that for consistent results in panel data analysis, the number of instrumental variables should be fewer than the number of groups (observations) He advised that System-GMM results should only be used when the panel data exhibits the characteristics of "small T, large N." Roodman illustrated that if the time period (T) is sufficiently large, dynamic panel bias becomes negligible Conversely, if the number of observed individuals (N) is not adequately large, the reliability of cluster-robust standard errors and the Arellano–Bond autocorrelation test may be compromised.
Table 4.6 System-GMM regression result for Step 1 - Model 1
Number of instruments < Number of groups Yes p-value AR (1) 0.012 p-value AR (2) 0.104
This table shows the regression results of Model 1 by System-GMM to determine the significant correlation between the outcome variable, firm value (FV) - lnFV, and the main explanatory variable,
In this study, cashflow volatility (CFV), represented as lnCFV, serves as the initial step in identifying the mediating variable based on the research framework established by Baron and Kenny (1986) Key control variables include profitability (Profit) as lnProfit, growth opportunity (MBVE) as lnMBVE, and tangibility (Tang) as lnTang, along with year dummies To address endogeneity issues among regressors, the model incorporates lagged values and first differences of the independent variables, following the methodology suggested by Roodman (2009) The analysis is conducted using data from 2018 onwards.
for 1%.
The Hansen test results indicate a valid model with a p-value of 0.066, which falls between 0.05 and 0.25, confirming stability The first autocorrelation shows significance with a p-value of 0.012, while the second autocorrelation is insignificant at 0.104, fulfilling the necessary conditions Additionally, the model employs fewer instrumental variables (13) than the number of groups (259), reinforcing the model's persistence and ensuring that the results are consistent and unbiased Consequently, these findings are reliable for further analysis.
The regression analysis conducted using the GMM model reveals that all variables are statistically significant, with p-values below 0.1, aligning with expected outcomes Notably, cashflow volatility (CFV) exhibits a negative relationship with firm value (FV), indicated by a correlation coefficient of -0.023 at a 5% significance level This suggests that a 1% increase in cashflow volatility results in a 0.023% decrease in firm value, assuming other factors remain constant.
The analysis reveals that a 1% increase in profitability (Profit) leads to a 0.026% increase in firm value (FV), indicating a significant relationship at a 10% significance level Additionally, growth opportunities (MBVE) show a strong positive correlation with firm value, as evidenced by a regression coefficient of 0.348 and a significance level of 1% Furthermore, tangibility (Tang), which reflects a company's ability to secure loans, also positively impacts firm value; specifically, a 1% increase in the ratio of fixed assets to total assets results in a 0.051% increase in firm value, with a confidence level of 95%.
The findings from the first model in step one reveal that cash flow volatility has a significantly negative effect on firm value, aligning with the initial criterion of Baron and Kenny's four-step research model for examining mediating variables.
(1986) Furthermore, the results are also in harmony with the results of Njuguna et al
(2022) and other studies and answers the sub-research question 1 and H1: Cashflow volatility has a significant impact on firm value Therefore, the research is going to continue to Step 2
4.5.2 REM with clustered standard errors for Step 2 - Model 2
To accomplish the dissertation's ultimate objective, it is essential to demonstrate that cash flow volatility (CFV) significantly influences corporate investment (INV) Consequently, Step 2 focuses on identifying a statistically significant relationship between these two variables The findings from the REM with clustered standard errors method for Model 2 are presented in the table below.
Table 4.7 REM with clustered standard errors regression result for Step 2 -
Model 2 Random-effects GLS Regression (Y = lnINV; X = lnCFV)
The regression results from Model 2, utilizing REM with clustered standard errors, reveal a statistically significant correlation between cashflow volatility (CFV) - lnCFV and corporate investment (INV) - lnINV This analysis represents step 2 in the four-step mediation detection framework established by Baron and Kenny (1986) The findings are based on a comprehensive dataset comprising numerous observations.
for 10%, 5%, and 1%, respectively.
CONCLUSIONS AND POLICY IMPLICATIONS
RESEARCH FINDINGS
This section addresses three sub-research questions and one key research question from Chapter 1 by analyzing 300 non-financial companies listed on the HOSE in Vietnam The findings are based on consistent and unbiased results obtained from regression models during the study period Additionally, this section highlights further discoveries and contributions made by the author throughout the research.
The dissertation's first step in the mediating variable research model provides practical evidence supporting traditional asset valuation models, which view volatility as a risk that negatively impacts company value, in contrast to option pricing theory, which sees volatility as enhancing equity value This finding aligns with the research of Minton and Schrand (1999), Beladi et al (2021), Rashid et al (2022), and Njuguna et al (2022), while contradicting the studies of Shipe (2015) and Gworo (2019) From an investor's perspective, cash flow fluctuations indicate a high risk level for the enterprise, suggesting that high operating cash flow volatility signals operational difficulties and instability in net incomes, increasing the risk of earnings for investors Consequently, potential investors may hesitate to invest, leading to an undervaluation of the company, while existing shareholders may question the effectiveness of operations, potentially resulting in withdrawal and decreased investment, both of which contribute to a decline in firm value.
In step 2, the author employs regression analysis of corporate investment against cash flow fluctuations using Robust Error Model (REM) with clustered standard errors, revealing a negative relationship between operating uncertainty—specifically cash flow volatility—and corporate investment This finding addresses sub-question 2 and supports the conclusions drawn by Minton and Shrand (1999) as well as Beladi et al.
(2021), Rashid et al (2022), Njuguna et al (2022), and opposites to the studies of Cohen
According to the Pecking Order Theory by Myers and Majluf (1984), businesses prefer internal funding sources like retained earnings and operational cash flow for investments, as external financing increases costs Unstable cash flow can significantly impact investment decisions, leading to two potential scenarios: companies may either cut back on investments due to cash shortages or opt for costly external funding options, such as debt and bonds, to maintain their investment activities despite cash flow fluctuations.
Investment activities are diminished from the enterprise's perspective due to both cases, highlighting the findings of Fazzari et al (1987), which indicate that investment expenditures are highly responsive to fluctuations in cash flow.
The findings from step 3 reveal a significant positive relationship between corporate investment and firm value, indicating that expenditures on capital, advertising, and R&D enhance business value, thus addressing sub-question 3 Capital expenditures serve to sustain current operations, while advertising and R&D are geared towards generating future cash flows (Richardson, 2006; Park & Jang, 2013) Consequently, increased corporate investment is perceived by investors as a strategy to boost revenue and future income potential, leading to an appreciation of the business and an increase in market value Park and Jang (2013) emphasized that the primary goal of investment activities is to improve market positioning and generate cash flows Overall, the positive correlation between investment and firm value aligns with the findings of Johnson and Pazderka (1993).
Ye and Yuan (2008), Kim et al (2018), Dang et al (2020), Mousa et al (2021), and Njuguna et al (2022)
This research aims to determine if corporate investment mediates the relationship between cash flow volatility and the firm value of non-financial companies listed on the HOSE Utilizing a four-step mediating variable model and regression methods, including System-GMM and REM with clustered standard errors, the study identifies significant evidence of corporate investment's mediating effect on the connection between cash flow fluctuations and corporate value These findings align with the conclusions of Njuguna et al.
A 2022 study of non-financial companies listed on the Nairobi Securities Exchange highlights the role of firm investment as a mediating variable, supported by Free Cash Flow Theory (Jensen, 1986), Underinvestment Theory (Myers, 1977), and Pecking Order Theory These theoretical frameworks provide insights into how investment decisions impact company performance and capital allocation strategies.
Significant fluctuations in cash flow can indicate periods of surplus or shortfall for a business (Myers & Majluf, 1984) According to Jensen's Free Cash Flow Theory, when companies experience excess cash flow, managers may invest in projects with negative Net Present Value (NPV), ultimately harming corporate value The Pecking Order Theory suggests that businesses prefer internal funding, but when cash flow is scarce, they may struggle to invest and might need to seek external financing through debt or new shares, which often comes at a higher cost This increased financing cost, driven by cash flow instability, signals a challenging business environment to investors, leading to further erosion of business value due to underinvestment (Jensen & Meckling, 1976; Myers & Majluf, 1984).
1984) Both cases show that cashflow fluctuations affect investment decisions, thereby causing a decrease in firm value
The study reveals four key findings: Firstly, investors in both emerging and developed economies tend to underestimate the impact of cash flow volatility on firm value, despite differing cultural and political contexts Secondly, heightened business uncertainty, particularly regarding operating cash flow, results in a decline in corporate investment across both developing and developed nations Thirdly, increased corporate investment can signal positive prospects to investors, thereby enhancing the company's market value Lastly, corporate investment serves as a crucial intermediary between cash flow fluctuations and the value of non-financial firms listed on HOSE, as it boosts enterprise value while cash flow volatility disrupts investment strategies, leading to underinvestment or overinvestment and a subsequent decrease in business value Additionally, during uncertain times, businesses often cut back on investment activities, further contributing to a decline in overall business value.
From the findings presented in Section 5.1.1, this study has provided four main contributions to the existing empirical research
This study highlights the intermediary role of investment in the relationship between cash flow fluctuations and the enterprise value of non-financial companies listed on the HOSE from 2018 to 2022 Additionally, it offers insights into the factors influencing enterprise value, specifically focusing on investment activities and cash flow variations, thereby laying the groundwork for future research in Vietnam.
The empirical findings on the relationship between cash flow fluctuations, investment, and enterprise value offer valuable insights that address inconsistencies in prior research regarding corporate responses to unpredictability Additionally, these results provide essential data for understanding the dynamics among these variables, highlighting the critical role of corporate investment in developing countries and small economies.
The dissertation presents a comprehensive analysis of the relationships among key variables in Vietnam, utilizing a robust dataset of 1,500 firm-year observations from 300 non-financial listed companies on the HOSE, covering the period from 2018 to 2022 The author asserts that the findings will significantly contribute to future research on related topics.
The dissertation outlines a suitable research methodology that incorporates various approaches, including REM with clustered standard errors and System-GMM, for future investigations of mediating factors in Vietnam, drawing on the Baron and Kenny (1986) model This is particularly significant as there remains a limited number of both official and unofficial studies addressing this topic in Vietnam.
Fluctuations in an enterprise's cash flow can arise from various external and internal factors, resulting in decreased revenue and increased expenses, which destabilize profits and operating cash flow Research indicates that these fluctuations can lead to reduced investment spending and a decline in business value Consequently, it is crucial for policymakers and business managers to recognize their role in minimizing operational risks and maximizing firm value.