Determinants of nonfinancial firm capital structure in UK

This research aims to employ longterm leverage and total leverage to investigate the capital structure determination of 200 sample FTSEL Allshare firms in manufacturing and service sector in the UK and to conduct a comparative research of capital structure determination across those two sectors. Such sectors have its own business environments as well as characteristics. Thus, the capital results from such sectors might vary. The study variables are developed based on prior literature and empirical evidence. Financial data are collected from FAME and annual reports. Statistical tests include correlation analysis, descriptive statistics and pooled OLS regression analysis.

Determinants of non-financial firms capital structure

Abstracts

This research aims to employ long-term leverage and total leverage to investigate the capitalstructure determination of 200 sample FTSEL All-share firms in manufacturing and servicesector in the UK and to conduct a comparative research of capital structure determination acrossthose two sectors Such sectors have its own business environments as well as characteristics.Thus, the capital results from such sectors might vary The study variables are developed based

on prior literature and empirical evidence Financial data are collected from FAME and annualreports Statistical tests include correlation analysis, descriptive statistics and pooled OLSregression analysis

Based on the empirical results, it can be seen that manufacturing sector normally borrow moredebt than their counterparts in two other sectors The descriptive analysis suggested that for long-term debt to equity ratio during 5 year-period from 2011 to 2015, manufacturing companiesprefer to use more long-term debt to finance for their investments which implies thatmanufacturing firms rely on more long-term debt than the companies in service sectors sincethey are capital-intensive companies Since manufacturing companies use more long-term debts,the service firms are concluded to have higher short-term leverage, suggesting that the capitalstructure of service firms in the UK tend to have higher level of short-term debt which aresuitable for the roll over of their expenditure and current investments for the working capital for

a year Higher asset tangibility and the size of the firm imply that manufacturing companies havehigher investments on non-current assets Since firms in manufacturing industry rely largely onlong-term financing, this leads to lower liquidity position as compared to the service companies.This empirical observation is in line with the trade off and agency argument for capital structurerelevancy Dividend payment in manufacturing firms tends to be lower than that of the servicecompanies, which can be explained by higher growth opportunities and large investment in fixedassets in manufacturing sector Result by performing pooled OLS multiple regression with paneldata, the model was free from the problem of multicollinearity and proved that this model iseconometrical robustness The trade-off theory (TOT) and pecking-order theory (POT) are thetwo main strands of theories and the testable hypotheses are built and tested based on them In

terms of dividend, it can be easily seen that there is a positive correlation between dividend anddebt levels in all two sectors, which provide strong evidence support for the POT and suggestingthat TOT had no effect on such variable No-debt tax shield and firm growth also support forPOT Nevertheless, in term of firm size and profitability, there is strong evidence for inverserelationships with long-term and total leverage ratio, supportive for POT With respect totangibility, the result was found to support for TOT in all proxies and for two sectors The resultfrom liquidity variable is complicated while long-term debt to equity support for TOT while totaldebt-to-equity support POT in two sectors.

Foremost, I would like to express my sincere thanks to my advisor Prof … for the continuous support of my Bachelor study and research, for his patience, motivation, enthusiasm, and massive knowledge His guidance helped me a lot for this dissertation

Besides my advisor, I would like to express my sincere gratitude the rest of my lecturers and tutors at University, for their encouragement, guidance, dedication during my academic

courses

Last but not the least; I would like to to convey my sincere thanks to my family: my parents, for giving birth to me at the first place and supporting me spiritually throughout the term year

Table of Contents

The capital structure theory was introduced by Modigliani and Miller (1963) While mostempirical studies argue that companies could benefit from tax deduction of using debt to financetheir operations which may positive affect firm value, no consensus study investigating thefactors affecting firm capital structure Literature review section provides conflicting findings ofthe determinants of firm capital structure In addition to this, few studies have examined thefactors influencing capital structure in the UK market One of the main purpose of this paper is tocontribute to the literature on the decision of the UK firm capital structure With respect to thecapital structure decision, a wide range of factors may influence the leverage level of a firm such

as shareholder, board characteristics, and firm-specific factors This paper focuses oninvestigating the impact of firm-specific factors including firm size, profitability, liquidity, non-debt tax shield, growth and tangibility on the UK firm capital structure proxied by firm leveragelevel Unfortunately, the majority of empirical studies investigate determinants of capitalstructure in the US market, only few studies test the capital structure theories on worldwidemarket For example, the worldwide empirical study of Wald (1999) provide evidence that theimpact of firm-specific factors on capital structure is different across countries and industries.This means that each country may have different characteristic leading to the difference firmcapital structure decision Using empirical data of the UK non-financial firms, this paper willfeedback the literature on the impact of firm-specific factors on level of leverage

When it comes to the capital structure theory (irrelevance proposition) of Modigliani and Miller(1958, 1963), some modern capital structure theories based on the MM theory are given underthe relaxation of some unrealistic assumption of the MM theory The static trade-off and peckingorder are two main theories that incorporate some improvements on the original capital structuretheory This paper takes into account both trade-off theory and pecking order theory to developresearch hypotheses Further details of these theories are given in the literature review section.1.2 Research Objectives

The main purpose of this study is to examine the determinants of the all UK non-financialcompanies over the period between 2000 and 2011 Based on the major theories relating to thedeterminants of firm capital structure, the study aim to contribute to the literature of capitalstructure in the UK economy The research objectives are detailed in the following bullet points

• To identify the factors affecting the capital structure decision of the UK firms

• To identify the strongest factors affecting the capital structure

• To give the feedbacks to the literature of capital structure studies

1.3 Research Questions

The paper aim is to answer the following questions

• What are the statistically significant factors having an impact on the non-financial firms

in the United Kingdom?

• Which are the strongest factors affecting the capital structure decision of the UK financial companies?

non-• Whether the findings of this paper are consistent with the recent studies of thedeterminants of capital structure?

1.4 Research methods overview

The purpose of this paper is to examine the impact of the UK non-financial firm’s characteristics

on the level of leverage In particular, firm characteristics include firm size, growth, profitability,non-debt tax shield, dividend and tangibility This paper uses non-financial companies in theFTSE All Share Index as the representation of the UK non-financial firms To reflect the currenttrend of the determinants of firm capital structure, this papers only examined firms during theperiod between 2011 and 2015 The main methodology of this paper is the Pool Ordinary LeastSquare (OLS), which could provide reliable results of the impact of firm-specific factors on thelevel of leverage

Literature review: this section focuses on reviewing recent findings of the determinants of capitalstructure and two main capital structure theories, namely trade-off theory and pecking theory.Based on the review of empirical studies on the impact of firm-specific factors on capitalstructure, hypotheses are also developed in this section

Methodology: This section provides the research method used for testing developed hypotheses

in section 2 (Literature review), and the model of this paper Data collection method and analysis

is also given in this section

Results: This section summarizes the finding of OLS regression models to determine the impact

of firm-specific factors on capital structure Results section also gives the comparative analysis

in three different sectors in terms of decision of capital structure

Conclusion: This section summarizes the main findings and contributions of this paper Also,some limitations and recommendation for future researches on the determinants of capitalstructure will briefly given in this section

2 LITERATURE REVIEW

This section provides a short review of major theories relating to the capital structure andpotential variables that are expected to have an impact on the firm capital structure Thehypotheses are given based on given theories

2.1 Modigliani and Miller capital structure theory

Under the perfect markets where taxes, bankruptcy costs and transactions costs do not exist andall information are disclosure, the level of debt in the firm capital structure does not affect thefirm’s market value (Modigliani and Miller, 1958) Because of the existence of taxes is inevitable

in every economy in the world, a new theory by Modigliani and Miller (1963) calledincorporates taxes and relaxes the assumption of no tax This leads to the realistic of theinvestigation of the impact of tax on the firm capital structure

There are different capital structure decisions that a company can make to finance their businessoperations Most companies tend to use both debt and equity to finance their operations On theother hand, many companies solely use equity for their operations and do not get involve in debt.There are different reasons that could explain the conflicting trends of capital structure The trend

of using debt is first developed by the capital structure theory of Modigliani and Millers (1958)which states that using debt does not affect firm market value This study has created afundamental background of capital structure theory However, the main weakness of this MMtheory is its unrealistic assumptions that the theory is only valid in perfect market where tax,brokerage fees, asymmetric information, transaction cost, bankruptcy cost are not existed, and all

investors have rational behavior Moreover, no investor or security issuer is large enough toinfluence the market in the perfect market Although these assumptions are unrealistic, thecontribution of the theory has been a cornerstone of capital structure theory Due to theunrealistic assumptions, many studies have relaxed some assumptions of MM theory to updatethe irrelevance theory The most significant update of Modigliani and Miller (1958) theory is theModigliani and Miller (1963) theory which relaxes the tax assumption, which means that the

MM (1963) theory takes into account the influence of tax on firm market value This updatedtheory has significantly contributed to the capital structure theory as tax is always existent andinevitable However, there are some incorporated assumption of the MM theory with tax Inparticular, no transaction costs exist and both individuals and corporates can borrow at the samerate The proposition II of MM theory suggests that as the firm’s level of leverage increases, theestimated return on equity will increase in linear due to the tax reduction of using leverage If afirm spends interests for their loans, the tax obligation will be lower, which in turn improvesfirm’s return on equity

To support the MM capital structure theory, the study of Miller (1977) investigates the impact ofprivate tax and corporate tax on capital structure However, the finding of Miller (1977) isconflicting with the finding of Modigliani and Miller as he argued that corporate tax could betrade off by the associated tax benefit of private taxes resulting in the irrelevance of tax oncapital structure decision In addition to this, the study of Fama and French (1998) finds that theadvantage of tax could lower firm market value, which is inconsistent with the previous capitalstructure theories of Modigliani and Miller (1963) and Miller (1977) The inconclusive of theimpact of tax on firm market value and capital structure decision would encourage future studies

on capital structure and irrelevant theory with more relaxation on unrealistic assumptions onbrokerage fees, transaction costs, bankruptcy costs and asymmetric information The followingsubsections will provide details about some major capital structure theories

2.2 Static trade off theory

The trade-off theory considers both bankruptcy costs and taxes based on the Modigliani andModel models, which products a benefit for company holding a high level of debt due to the taxshield In other words, this theory encourages firms to balance the level of debt and equitybecause these can play lower taxes due the tax deductible from debt (Jong, Verbeek and

Verwijmeren, 2011) Using both equity and debt is a common trend of firms in the world as theycan benefit from tax deduction and allow them to control the risk of using leverage However,determining the optimal ratio of equity to debt is the most important question The main purpose

of capital structure decision is to find the optimal ratio of debt and equity which enhances firmmarket value and shareholder value Kjellam and Hansen (1995) argue that a proper capitalstructure decision allows a company to effectively combine financing sources to maximize firmmarket value and minimize the cost of capital Unfortunately, it would be impossible to find anoptimal point based on capital structure theories due to the unrealistic of their assumptions of aperfect market where transaction, agency, brokerage and bankruptcy costs do not exist To applythe capital structure theory into the practice, it is essence to relax some assumptions of a perfectmarket In reality, there is a link between debt and the insolvency risk that could associate withbankruptcy costs The trade-off theory takes into account bankruptcy costs in determining theoptimal capital structure The theory was introduced by Myers (1977) and argues that there arebenefits of using leverage within a capital structure up until the optimal capital structure isreached Using the mix structure of debt and equity to balance the costs and benefits is the mainpurpose of the trade-off capital structure theory The trade-off theory also takes into account thebenefit from interest payments, financial distress costs, agency costs of debt and agency benefits

of debt Welch (2004) introduced the following model of trade-off capital structure theory:

Where VL stands for the market value of leveraged firm, VU stands for unleveraged firm value,

PV stands for the interest tax shield, which indicates the benefit of tax deduction of usingleverage, PV (Agency Benefits of Debt) is the present value of agency benefits of debt,PV(Financial Distress Costs) is the present value of financial distress costs arising when acompany employ excessive the optimal level, PV (Agency Costs of Debt) and PV (Agency Costs

of Debt) refer the present value of agency issues

2.3 Agency theory

Freedom of agency cost is an assumption of the Modigliani and Miller irrelevance theory in aperfect market Jensen and Meckling (1976) developed agency theory that relaxes the agency

cost assumption of MM theory The agency cost arises when the interests of a company’sshareholders conflict with interests of its managers as the objectives of the two parties might bedifferent On the one hand, firm’s managers with their deep understanding on business operationand capital structure could improperly use their power to get involve in potential projects withnegative NPV to maximize their wealth and power In addition to this,

On the other hand, shareholders require firm’s managers to run the company in a way thatmaximizes shareholder value Jensen and Meckling (1976) argue that the managers and agentscan exploit their power to generate personal benefit which negatively affect firm market valueand shareholders’ interests This leads to the high separation in a firm operation and high agencycosts Agency issues tend to arise in listed companies where shareholders have rights to influencecompany’s operations The conflict of interests between managers and shareholders would result

in high agency costs to control the managers With respect to the relationship between agencycosts and capital structure, Kjellam and Hansen (1995), Fama and French (2002) and Berger andPatti (2006) find a statistically and negatively relationship between these factors For example,these authors argue that the relatively high level of free cash that are available for managerswould induce them to overinvest in negative NPV projects that could increases their wealth andpower which negative influence shareholder value and firm market value In other words, there is

a negative association between firm leverage proxied by the ratio of debt to asset ratio and theagency costs proxied by the ratio of general administration costs to firm total assets Thisindicates that using debt to finance firm operation is a method to lower agency cost or conflict ofinterest To support this argument, Frank and Goyal (2007) find that using leverage would allowfirms to lower their free cash flows due to the interest payments Lowering available cash flowwould obstruct firms’ managers to overinvest or exploit to generate personal benefits.Furthermore, using leverage could also improve private monitoring as creditors have strongincentives to observe and create pressure on firms’ managers to increase firm profitability(Huang and Song, 2006)

2.4 Pecking order theory

Although the trade-off theory has been a cornerstone of the capital structure theory, it has failed

to explain the trends that the yield of stock price tends to increase and decrease with theleverage-increasing and leverage-decreasing volumes, respectively (Chen, 2004) Pecking order

theory explains that companies tend to prioritize their internal financing, then debt and the lastresort is equity This means that equity is less preferred than debt to raise capital because whenthe managers raise capital by equity, investors tend to think that the firm value is overvalued andthe managers are exploiting the over-valuation even though the managers have betterunderstanding about the company than investors (Myers and Majluf, 1984) Under the peckingorder theory, the primary source of financing is retained earnings (internal sources), the nextchoice is long-term debt, and follows by issuing shares (equity) This explains why profitablecompanies prefer internal financing and are less influenced by debt or low leverage level.

Academic suggest that pecking order theory is alternative view of trade-off theory in terms ofdetermining the ratio of debt to equity Frank and Goyal (2007) argue that while trade-off capitalstructure theory focuses on determining an optimal point of debt to equity, pecking order theorydoes focus on the need of external financing, which determines the level of leverage of acompany For example, firms with high profitability tend to finance their operations by internalfunds or low level of leverage and are less dependent on external funds

2.5 Capital structure strands of theory and research gap

It is clear from previous sections of capital structure theories that both static trade-off andpecking order theory relax different assumptions of MM theory In particular, trade-off theoryallows firm’s managers to determine an optimal debt to equity ratio, while the pecking ordertheory focuses on the choice of firm financing based on their need of external funds andprioritizes the internal sources Many empirical studies of the trade-off theory and pecking ordertheory on capital structure decision provide inconclusive results in the UK market The mainpurpose of this paper is to develop hypotheses based on two theories and test whether thosehypotheses are valid in the UK market The gap of these theories will be discussed in the nextsection Most empirical studies provide inconclusive findings of the relationship between capitalstructure and the trade-off theory and pecking order theory and examine the relationship in singlepaper rather than combine them in the UK market The finding of this paper is expected tocontribute to the literature in this direction

2.6 Dependent and Independent variables

2.6.1 Dependent variables

Based on the empirical study of Noulas and Genimakis (2011), the dependent variable of thispaper is the level of leverage which represents firm capital structure Two dependent variables ofthis paper are the calculated by the long-term debt to equity and by the firm total debt to equity.L1 stands for the ratio of long-term debt to equity, which indicates the long-term financingsources invested in firm assets (Akintoye, 2009) This explain the reason why this paper takesinto account both long-term debt and total debt However, short-term debt would not beappropriate to determine the firm level of capital structure as it may not reflect characteristic offirm capital structure Using long-term debt and total debt to equity would not conflict withcapital structure theories

2.6.2 Independent variables

The main purpose of this paper is to investigate the impact of firm-specific factors on capitalstructure theories In particular, firm-specific factors include firm size (SIZE), profitability(PRFT), growth (GROW), Tangibility (TANG), liquidity (LQDT), non-debt tax shield (NDTS)and dividend (DIVD) Details of the firm-specific calculations are given in section 3.5.2

• (SIZE) Size of company

According to Driffield et al (2007), there is a strong relationship between firm size and debtlevel of company Smaller firms might suffer higher of bankruptcy costs than its counterparts.Sharing this opinion, Fulghieri and Suominen (2012) discuss that owing to be more diversified,larger firms might fail less often This hypothesis is proven by number of empirical test.Fulghieri and Suominen (2012) use pooled regression to test the relationship between firm sizeand level of debt Their result suggests that firm size is positively associated with debt debt leveland it associated with trade-off theory Furthermore, King and Santor (2008) discuss that largerfirms have quite low problem of asymmetric information with capital market because they areclosely watched by publics than smaller ones Vithessonthi and Tongurai (2015) find that largecompanies tend to use long-term debt for their future strategy and invest in potential projects,while smaller firms tend to use short-term debt to finance their operations and have a low long-term debt This means that size is positively associated with leverage Fama and Jensen (1983)document that while large firms tend to provide enough financial data to lenders to get long-term

fund, small firms tend to not fully provide their financial statements to lenders resulting in alower chance to get long-term fund This leads to the trend the larger firms tend to have higherlevel of leverage than smaller ones Thus, the hypothesis can be written as follows:

H1a: As stated by trade-off theory, it is expected that firm size is positively related to leverage However, some scholars reject the above hypotheses The empirical evidence of Marchica andMura (2010) purport that big company can accumulate huge number of retained earnings theirtests show that there is a negative link between firm size and debt level Agree with thisviewpoint, Sheikh and Wang (2011) suggest that larger firms are less likely to use leverageowing to their strong cash flows Aivazian, Ge and Qiu (2005) argue that large firms tend toprovide more information to the external investors than smaller ones However, large publishedfirms with low problem of asymmetric information tend to finance their operation and strategy

by equity rather than debt This results in a trend of low level of leverage of large firms All ofwhich, support a hypothesis of a negative relationship between leverage and firm size This ideasupport for theory of pecking order and proposing for the author the below hypotheses:

H1b: As stated by pecking-order theory, it is expected that firm size is negatively related toleverage

• (PRFT) Profitability of company

Firm profitability is one important attribute of capital structure There are number of scholarssupporting for trade off theory, suggest that bankruptcy costs, agency costs and taxes leading tohigher book leverage of firm Actually, Ellul (2010), suggesting that there is a negativerelationship between profitability and business bankruptcy Agency cost model also suggest thatthe higher leverage result in decreasing of agency issues (La Rocca 2007) Sheikh and Wang(2011) document that high profitable companies tend to have low insolvency risk which mayinduce them to take more debt This leads to the high level of leverage in profitable companies.These arguments support the hypothesis of a positive relationship between profitability and level

of leverage Thus, the hypothesis can be drawn as follows:

H2a: As stated by trade-off theory, it is expected that profitability is positively related to leverage

Alternatively, according to pecking-order theory, higher profitability of company results in lowleverage According to the empirical work of Nielsen et al (2006), there is a significant negativeassociation between leverage and profitability The firm with high level of retained earnings ismore likely to use their internal fund to finance their investment rather than using externalsources (Lemmon and Zender, 2010) In addition to this, Noulas and Genimakis (2011)document that a high level of leverage could lead to a high problem of free cash flow and maylower bank profitability The author support a negative relation between profitability andleverage This coupled with these arguments above lend support to the hypothesis of a negativerelationship between leverage and profitability Thus, the alternative hypothesis can be suggested

as follows:

H2b: As stated by pecking-order theory, it is expected that profitability is negatively related toleverage

• (TANG) Tangibility of company

According to previous study about the relationship between tangibility and leverage by Farhat et

al (2006), Caprio and Croci (2011), there is a positive relationship between leverage andtangibility The ratio of fixed to total assets is the major attributes in describing the capitalstructure decision in the UK companies The trade-off theory suggested that tangibility ispositively related to leverage because it can minimize the issues of conflict of interest betweenshareholder and bondholder, for example, collateral (Antoniou et al 2008) If the debtor putcollateral as one part of contract, the creditor is more confidence of the money since they thinkthat their debt is more likely to be paid In the worst circumstance, company assets still keepmore valuable in liquidation (Daskalakis and Psillaki 2008) Therefore, the firm with moretangible assets seems more confident in raising leverage in their capital structure owing toincreasing its guarantee of repayment Thus, the asymmetric information issues, which arementioned by trade-off and pecking order theory, will be minimized The hypotheses can bewritten as follows:

H3: As stated by trade-off theory and pecking-order theory, it is expected that tangibility ispositively related to leverage

• (GROW) Growth rate of company

According to Huang and Song (2006), who support for trade-off theory, suggests that highgrowth rate firm seems more likely to meet business bankruptcy since they are more likely to addmore debt on their capital structure The firm with higher growth rate can increase theircompany’s value by assessed to have capital assets However, those assets are collateralized asfixed assets According to empirical work of Cohen and Yagil (2010), growth rate is negativelyrelated to leverage Myers and Majluf (1984) argue that firms with high opportunity to growthtend to finance their operating by equity These arguments lend support to the hypothesis of anegative association between firm growth and level of leverage Thus, the hypothesis is asfollows:

H4a: As stated by trade-off theory, it is expected that firm growth is negatively related toleverage

Alternatively, the pecking order theory purport different viewpoint Bie and Hann (2004), whosupport for this theory, suggest that there is a positive relationship between firm growth andleverage in firm capital structure According to them, the debt level will increase until investmentopportunities surplus the ability of generating funds internally and vice versa Thus, the higherfirm growth might lead to the higher debt level Jensen and Uhl (2008) argue that although firmwith high potential growth tend to prioritize retained earnings, they are not enough to allow firm

to follow potential projects This encourage those firms to take more debt for their futurestrategy, which support the hypothesis of a positive relationship between firm growth and level

of leverage Thus, the hypothesis can be proposed as follow:

H4b: As stated by pecking-order theory, it is expected that firm growth is positively related toleverage

• (LQDT) Liquidity of company

There are number of empirical works research about the association between leverage andliquidity However, the results seem vary Fulghieri and Suominen (2012) document that highlevel of leverage is associated with high liquidity as debts could be used to meet firm short-termobligations especially during the financial crisis period According to Al‐Najjar and Hussainey(2009), there is a positive relationship between leverage and liquidity ratio The firms wouldhave more ability to pay their short-term debt if their liquidity is increased and thus, minimize

the financial distress Thus, the company can keep their financing mix below the optimal point ofleverage This argument matches with trade-off theory Therefore, the hypothesis is as follows:H5a: As stated by trade-off theory, it is expected that liquidity is positively related to leverageAlternatively, according to pecking order theory, the company prefer use internal funds ratherthan external financial sources, thus, higher liquidity refers to the company have enough capitals

to use for their investments (Chen et al 2004) Hence, they are likely less relying on debt Firmwith highly liquidity level tend to have low level of leverage as they do not incur high interestpayments (Cohen and Yagil, 2010) Moreover, Noulas and Genimakis (2011) argue that highlevel of liquidity or cash flow would induce firm’s managers to overinvest, which do need moreexternal funds to follow potential projects This explains why many liquid firms have a highlevel of leverage Based on these arguments, it can be argued that firms with high liquidity tend

to have low level of leverage Thus, the hypothesis can be seen as follows:

H5b: As stated by pecking-order theory, it is expected that liquidity is negatively related toleverage

• (NDTS) Non-debt tax shield

Follwing Uhl and Jensen (2008), non-debt tax shield is computed by the ratio of depreciation tototal assets Interest payment is not only tax deductible but also depreciation Both Taxdeductible form debt and depreciation are taken into account to determine an appropriate firmcapital structure As stated by trade-off theory, the company with higher profitability is morelikely to have higher tax burden, as a result, improving tax advantage of debt (Anderson et al2010) Correspondingly, Caprio and Croci (2011) and Croci et al (2011) also suggest that interestexpenses can be seen as perfect alternate for non-debt tax shield They find that there is apositive association between non-debt tax shield and leverage by regressing the leverage forlong-term debt to equity and total debt to equity ratio against the non-debt tax shield (Goh 2005).The tax is saved from interest expense can be seen as replacement for tax advantage in case offirm use debt as the main financial source Marchica and Mura (2010) find that firms tend toprefer external debt as it would be a signal of confident evaluation Moreover, high level of non-debt tax shield from leverage allows firms to lower income tax on their profit before tax

The hypothesis can be written as follows:

H6: As stated by trade-off theory, it is expected that non-debt tax shield is positively related toleverage

• (DIVD) Dividend

According to the empirical work of Lemmon and Zender (2010), the dividend payout ratio is animportant factor to minimize the agency conflict of interest Actually, this practice is an effectivecorporate governance mechanism to monitor and control the managerial behaviors which asgreed by Mallikarjunappa and Goveas (2007) and Daskalakis and Psillaki (2008), make interest

of shareholders and managers become more aligned The hypothesis can be written as follows: H7a: As stated by trade-off theory, it is expected that dividend is negatively related to leverage

On the other hand, the pecking order theory proposes that dividend is an important monitoringtool to mitigate the asymmetric information issues (Sayılgan et al 2006) According to Molly et

al (2012) the company might lower their free cash flow for their reinvestment decisions when ittends to keep high payout ratio Thus, in this situation, the firm is obliged to borrow money fromexternal financing sources, which might lead to increase leverage into firms’ capital structure(Villalonga and Amit 2006) Returning cash to shareholder is regular practice of public firms inthe UK thus, non-financial companies are unwilling to cut their dividend Thus, firms wouldborrow more debt to generate enough cash to return to their shareholders at fixed dates lead tothe situation that firm free cash flow are low (Gaud et al 2005) Thus, the alternative hypothesis

is as follows:

H7b: As stated by pecking-order theory, it is expected that dividend is positively related toleverage

2.6.3 Hypotheses

This section reviews all capital structure hypotheses that will be tested in the section 4

H1a: As stated by trade-off theory, it is expected that firm size is positively related to leverage H1b: As stated by pecking-order theory, it is expected that firm size is negatively related toleverage

H2a: As stated by trade-off theory, it is expected that profitability is positively related to leverageH2b: As stated by pecking-order theory, it is expected that profitability is negatively related toleverage

H3: As stated by trade-off theory and pecking-order theory, it is expected that tangibility ispositively related to leverage

H4a: As stated by trade-off theory, it is expected that firm growth is negatively related toleverage

H4b: As stated by pecking-order theory, it is expected that firm growth is positively related toleverage

H5a: As stated by trade-off theory, it is expected that liquidity is positively related to leverageH5b: As stated by pecking-order theory, it is expected that liquidity is negatively related toleverage

H6: As stated by trade-off theory, it is expected that non-debt tax shield is positively related toleverage

H7a: As stated by trade-off theory, it is expected that dividend is negatively related to leverageH7b: As stated by pecking-order theory, it is expected that dividend is positively related toleverage

3 METHODOLOGY AND DATA

3.1 Research philosophy

Following the empirical study of Noulas and Genimakis (2011), this paper employs quantitativemethod with secondary data to investigate the determinants of firm capital structure on non-financial firms in the UK The secondary data is collected from firm financial statements,surveys and the calculation of the author

The main purpose of this paper is to investigate the impact of firm-specific factor on the capitalstructure proxied by firm level of leverage on the UK non-financial firms The procedure ofanalyzing regression results of this paper is summarized by three following steps

Hypotheses: Based on the literature of firm-specific factors and firm level of leverage, main

hypotheses are developed and summarized in section 2.6.3

Hypotheses testing: Using OLS regression model to test whether mentioned hypotheses are

valid in the UK market

Results: Based on the empirical finding, this paper will feedback the literature on the association

between firm characteristic and capital structure and capital structure theories

3.2 Research strategy

The main core research strategy of this paper is archival research Saunders et al (2012) arguethat this strategy allows financial data and academic data to be the principal sources Thefinancial data of this paper is secondary as it allows to author to save budget and time for theanalysis section The sample population of this paper is also large enough to reflect thecharacteristic of the non-financial firms in the UK

As dependent on the secondary data, the sources should be reliable and free of ethical problems.Two chosen sources for financial data are Fame and DataStream, which meet the criteria in terms

of reliability, quality and non-ethical issues With respect to the academic sources, a wide range

of journals in the references section are collected from high-quality and reliable sources such asScience Direct, Emerald and Willey Online Library to develop the study hypotheses and reviewrecent findings

In short, there are two main kinds of data sources compiled in this paper The academic sourcesused to review recent findings of the determinants of capital structure and capital structuretheories Based on this, appropriate hypotheses are mentioned to determine whether proposedhypotheses are valid in the UK market The financial data sources are used to compiled financialdata of the UK non-financial firms

Academic sources: The academic articles on the determinants of capital structure are collected

via online libraries like Science Direct and Willey Online Library These articles are chosen fromhigh ranking journals and reliable for references These high-quality sources are free from anyethical issues This allows the paper to construct proper and reliable theoretical framework anddevelop capital structure hypotheses

Financial data: is collected from two reliable sources, namely Fame and DataStream The

financial data is mainly compiled from published financial statements The data is reliable andfree of ethical problem

3.3 Data

The purpose of this paper is to investigate the impact of firm-specific factors on the level ofleverage in the UK over the period between 2011 and 2015 Secondary data is compiled fromFame and DataStream, which collect from published financial statements, surveys andcomputation of the author over the 2011-2015 period However, this paper only examines UKnon-financial firms as the leverage of financial firms such as insurance companies and banksmay not reflect their capital structure decision The list of non-financial firms bases on FTSE All-share of London Stock Exchange All chosen companies have enough financial data for thewhole examined period Although there are 753 companies in the FTSE All-share, only 321 non-financial firms meet the criteria as many companies do not have enough financial data or fall into

financial company group The sample population of this paper will examine randomly 200 firmswhich covers more than 60% of the population The large sample population is sufficient toreflect characteristic of the whole population.

3.4 Statistical analysis

3.4.1 Statistical analysis model

This research follows the empirical work of Noulas and Genimakis (2011), who use the pooledOLS regression model to address the research objective as to explore the capital structuredetermination and to provide sectoral analysis of 2 chosen industries of the capital structure inthe context of the UK The two proxies represent the leverage include long-term and total debt-to-equity Such two proxies will be regressed against 7 financial characteristics as set in previouschapter According to Gujarati (2003), the application of panel data will improve the econometricestimates and reduce the effect of model misspecification

The research model of this research can be performed as follows:

In this formula, J refer to two proxies (L1) and (L2) L refers to sample firms, t refers to theresearch period used in this research α mentioned the constant term, β means the coefficient ofpooled regression, ε denotes the disturbance term, SIZE stands for firm size, PRFT stands forfirm profitability, TANG stands for firm tangibility, GROW stands for firm growth, LIQD standsfor firm liquidity, NDTS stands for firm non-debt tax shield, and DIVD stands for firm dividend

3.4.2 Explanation of the model employed

This study implements independent pooled regression panel analysis to address the researchobjectives as examine the capital structure determination in the context of UK firms According

to Antoniou et al (2008), this approach will work well when the pooled are relatively similar.There are two different industries in this research, namely manufacturing and services Thechosen companies in each group are fairly similar and homogenous Thus, the pooled OLSregression panel seems the appropriate choice in this situation There are 200 companies to beselected in this research over the 5 years period (Abor and Biekpe 2005) Thus, by applyingpanel data analysis there are 1000 variables to be generated in this research and it allows the

author to announce that the empirical research can be appropriate and sufficient enough to makeaccurate conclusion about capital structure determination in the UK market.

3.4.3 Evaluation of the model employed

Implementing independent pooled panel can bring some advantages At first, this approach canefficiency manages the individual heterogeneity, unlike cross-sectional analysis By applyingindependent pooled panel, the risk of getting biased results, which appear from heterogeneityissue, is minimized The independent pooled panel data allow the researcher to generate largernumber of observation, leading to generate more accurate conclusions of the trend and pattern of

data (Mátyás 2007, Saunders et al 2012) However, the investigator also need to acknowledge

that using panel data is fairly complicated owing to issues of the distortion of the measurementerrors from data collection and model design Thus, the previous research generally usingrandom and fixed effects model to perform panel data analysis

3.4.4 Quantitative testing methods

In order to follow study objective, the researcher tend to use number of econometrical tests inthis study The SPSS, the famous quantitative software package, will be used in this research.The research decides to use this application since it is easy to use with user-friendly interface.This application is recommended by number of research in the field of business research

There are number of econometrical tests are used in this study At first, the descriptive statistics will be used to describe and compare variables numerically Secondly, the Pearson’s correlation

analysis will be conducted to examine the number of relationship between two varibles.According to Saunders et al (2012) +1 os perfect positive and -1 is strongest negative 0 can be

seen as perfect independence Thirdly, Multicollinearity analysis will be used to to identity

statistical problem of any two variables which have more than 70% of correlation According to

Cohen and Yagil (2010), high correlation between two variables can affect largely to regression

results Finally, the significant testing will be performed to test the degree of relationship

between variables The hypotheses will be tested at two-tails at the significance level of 5% Ifthe sample statistics in non- rejection region, the hypothesis will be accepted Otherwise, the null

hypothesis will be rejected (Saunders et al 2012):

FIGURE 1: HYPOTHESIS TESTING CRITERIA

3.5 Measurement of variables

This part will review all study variables in this research They are divided to dependent variablesand independent variables They are as follows:

3.5.1 Measurement of dependent variables

Following the work produced by Daskalakis and Psillaki (2008) and Antoniou et al (2008), thepresent empirical research employs two proxies to measure the dependent variables, including:

• The firm’s long-run leverage (L1) is measured by the book value of long-term debt andliabilities over the value of company’s equity

• The total leverage (L2 ) is measured at book balance sheet value of the firm’s total debtand liabilities over the value of equity of company

3.5.2 Measurement of undependent variables

Firm size

According to Fulghieri and Suominen (2012), size of firms is calculated by the natural logarithm

of firm’s total assets The data is gathered for Fame database sources