Academic journal article Asia-Pacific Business Review

The Dynamics of Capital Structure

Academic journal article Asia-Pacific Business Review

The Dynamics of Capital Structure

Article excerpt

Introduction

Most of UK studies have been concerned with cross-sectional variations in gearing (see Bennett and Donnelly (1993), Rajan and Zingales (1995), Bevan and Danbolt (2002). With the exception of Ozkan (2001), there is no other UK study that has focused on the dynamics of capital structure adjustments. In addition to the dearth of empirical works on capital structure dynamics in the UK, previous work in this area carried out elsewhere, give contradicting results. For example, while both Bradley et al (1984) and Titman and Wessels (1988) do not find evidence to support the theory of substitutability between non-debt and debt tax shields which is advocated by De Angelo and Masulis (1980), Givoly et al (1992), and Chiarella et al (1992) find that there is a substitution effect between debt and non-debt tax shields. Generally the results of a number of U.S studies like Bradley et al, (1984) and Titman and Wessels (1988) report evidence of a negative relation between earnings volatility and gearing. To the contrary, a UK study, Bennett and Donnely (1993), document a positive relation. See Harris and Raviv (1991) for more contradictions, few of which have been resolved so far. This study first tests a broader set of attributes which affect capital structure decisions. Further, this study uses an alternative methodology, Structural Equation Modelling (SEM) to test these attributes in a dynamic contest. A synthesis of the theory is provided and then taken further by empirically examining capital structure dynamics. Two different methodologies are used on the same sample in order to assess whether the contradictory results could be explained by the use of different methodologies. The analysis of the dynamics of capital structure starts with tests to establish whether firms adjust their capital structure towards an optimal ratio. The target capital structure adjustment models incorporate determinants such as tangibility, non-debt tax shields, growth opportunities, firm size, volatility of earnings, past profitability, current profitability, probability of bankruptcy, cash holdings, interest rates and corporate taxes.

The selection of SEM, as an alternative methodology has been prompted by the two previous empirical studies of the determinants of capital structure have used SEM. Titman and Wessels (1988), the pioneers of the use of structural equation modelling technique called Linear Structural Relationship (LISREL) in capital structure studies in the US, and Chiarella et al (1992), who subsequently employed a similar approach in a study of capital structure of Australian companies. These studies have claimed that LISREL estimation technique has a number of advantages over the conventional (standard) or traditional regression models, including its ability to recognise and mitigate measurement and specification errors, which have plagued regression based studies. Despite these statistical advantages, the results of these two studies that used LISREL technique failed to resolve the empirical contradictions and even generated more contradictions and perverse results than most studies that have used variants of the conventional regression estimation models. Before we can judge the practical contribution of the SEM model, we need to use it alongside the conventional regression estimation technique with the same data set so as to be able to compare their explanatory power against both the theory and previous empirical findings. This study is probably the first to provide evidence of the use of structural equation modelling technique on UK capital structure and in particular the examination of capital structure dynamics.

Having identified more appropriate proxies for the theoretical attributes in this study, both methods are being used on the same data set, one after another, and by using as far as the models permit, the same variables. The results of this study provide future researchers with an input into their decision as to which methodology to adopt in similar empirical investigations. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.