All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser. Open Advanced Search. DeepDyve requires Javascript to function. Please enable Javascript on your browser to continue. Capital structure and financing of SMEs: Australian evidence Capital structure and financing of SMEs: Australian evidence Cassar, Gavin; Holmes, Scott This paper investigates the determinants of capital structure and use of financing for small and medium sized enterprises.
Read Article. Download PDF. Share Full Text for Free. Web of Science. Let us know here. System error. Please try again!
In contrast, most SMEs do not have access to quality labor force, which normally translates to l productivity and uncompetitive, and in the long run affects the national growth and development. It is therefore not an act of accident to see several government and other institutions support SMEs train and maintain quality workforce.
The task before this researcher is to assess the impact of institutional support in the provision of managerial capacity building on the growth of selected SMEs in Ghana.
Using firm level data from SMEs from the ten regional capitals of Ghana, the results indicated that the provision of institutional support in the training of procurement personnel, bookkeeping and accounting and business plan preparation experts had positive impact on the growth and development of the SMEs.
Related Articles:. Home References Article citations. As no information is available about the parent, in particular its capital structure, those firms were excluded. Finally, to avoid problems with negative values and negative-equity firms, which can appear as a consequence of using book values rather than market values, all firms with over per cent leverage were eliminated. After the selection criteria were employed a final sample of 1, firms was available for analysis.
The ambiguous nature of the shareholders classification creates the potential for noise if these firms are included in the final analyses. The exclusion of these firms also avoids problems associated with the classification of inside and outside financing. All tests were re-run separ- ately without the exclusion of these firms with shareholders and firms with parent financing, none of the findings were significantly affected.
These variables were either taken directly from the survey or created using the sum of different financing options listed in the study. A summary of the different financing choices available on the survey and the variables developed from them is contained in Appendix A. Leverage LEV is the total debt of the firm divided by the total assets. Given the scale of the firms in the sample and the financial information available, market values of debt or equity could not be considered, as they generally are in studies examining larger listed firms.
Long-term leverage LONG is included, as not all components of leverage are homogenous. Apart from the obvious maturity and duration differences, long-term leverage is more fixed and arguably more deliberate, with greater contractual obligations and screening processes required.
The proportion of non-current debt was explicitly requested by the survey. Therefore short-term leverage SHOR is the difference between leverage and long-term leverage. The creation of an outside financing OUT measure provided an alternative but intuitive means of examining the influence of agency cost arguments upon financing choice. Consequently, issues relating to moral hazard and information asymmetry may be better examined through the use of outside finance. The use of an outside financing variable is consistent with recent re- search examining smaller firm financing Fluck et al.
In addition, although not used for the testing of hypotheses, inside financing was also created for descriptive purposes. In addition to outside financing, a bank financing BANK measure was also incorporated into the empirical testing.
Several studies have looked specifically at the use and level of bank financing by firms, although bank financing has received more attention when the focus of the research is on SMEs, rather than larger firms, due to the accessibility of this type of financing to SMEs Freed- man and Godwin, ; Storey, ; Cressy, Potential differences that might be found include the issuance of more financing for firms with particular characteristics, such as a potentially greater reliance of assets in place for securing debt.
Under the financing categories of outside financing and bank financing adopted by this study, whether the capital is provided as a loan or as equity is not considered relevant. Hence these variables overcome potential problems associated with the identification of debt and equity or the use of quasi-equity by SMEs Ang, The main consequence of using book values as opposed to market values is a lower positive coefficient between firm profitability and the financing variables.
The lower coefficient results from book values being unable to capture the positive correlation between market value of equity and the profitability of the firm. However, the applicability of market values as applied to SMEs can be ques- tioned. The use of market values, assumes that a market for firm equity exists. In many cases due to information asymmetries and agency considerations SMEs do not have such markets, or they are very illiquid.
In addition, given the influence of the major decision maker, the ability to value such equity, given a potential sale of firm equity is problematic. For example, the firm value with or without the founder managing the firm can vary substantially. This suggests that the applicab- ility of market values in regard to smaller firm capital structure is problematic.
Tests for the continuous variables, represented by the proportion of the dependent variable to total assets were undertaken using ordinary least squares OLS regression. Given the panel nature of the data there may be an argument to support the use of panel techniques for analysis, however, the lack of data points over time four suggested that employing a static framework was more appropriate. In particular, the limited data points over time would most likely lead to high standard errors on such a panel analysis, resulting in a relatively low powered test.
In earlier years where such information was available, the correlation between tangible non-current assets divided by total assets and all non-current assets divided by total assets was between 0. This suggests that the use of NONA as an instrument for fixed tangible assets is appropriate provided this variable is uncorrelated with the disturbance in the estimation model. In particular, ROA was bounded between 0. All the independent variables employed by this study have been used by previous empirical studies examining SMEs financing choice.
Results 5. Descriptive statistics Table 1 provides a summary of the descriptive statistics of the dependent and independent variables. The mean median leverage of the sample firms was 0. Including such firms would increase the magnitude of leverage. The mean long-term leverage suggests that it represents around 17 per cent of the capital of SMEs. The mean median short-term leverage of the sample firms was 0.
Outsider financing appears to constitute 38 per cent of the capital of SMEs. We estimate that insider sources constitute 52 per cent of the capital of the firms, with the remaining proportion represented through provisions for liabilities, deposits, outstanding claims, and investments by non-director employees which was classified as neither outside or inside financing. The correlations between the dependent variables and independent variables are provided in Table 2.
Examining the univariate relationships between the dependent and independent variables, of particular note is the con- sistent positive negative correlation between GROW ROA and the dependent variables. Strong correlations between NONA and the financing variables are also observed, however, the direction varies with the dependent variable exam- ined. In particular, there is a negative relationship between NONA and LEV, and NONA and SHOR, however, there is a significant positive correlation between 4 Alternative methods for adjusting the independent variables to reduce the influence of outliers were also investigated, including log transformation, squared and root transforma- tions, winsorisation and bounding values.
These other methods generally did not reduce the distributional problems associated with the variables. Results of analysis performed with unadjusted dependent variables were generally similar, except for some loss of precision increased standard errors. In addition, analyses were also undertaken varying the bounds or winsorisation with limited effect on results. Independent variables are three year averages based on to values.
Finally, examining the correlations between size and the other independent variables it is found that larger firms have a lower proportion of non-current assets, lower profitability and less risk. The predicted sign for the coefficients with respect to leverage is displayed under each independent vari- able. The results of the OLS regression between the five dependent variables and the independent variables with varying controls for industry are reported in Table 3.
Evidence from SME studies is mixed as to whether industry member- ship results in differences within the firms financing and capital structure Scherr et al. The problems with drawing inferences from these studies, include the varying use of other control variables between studies and the different proxy for industry effects employed.
Consequentially, industry may be proxy- ing the securability of assets or asset risk, depending upon the choice of vari- ables included in the empirical models. To ensure that the findings are not significantly affected by industry influences the multivariate analyses in Table 3 include tests both without industry controls and with industry controls, using one-digit ANZSIC groupings.
As shown, the control of industry group in the regressions had limited effect on the inferences found, although there was some limited support that including industry effects significantly increased the explanatory power of the model.
Given the limited influence of industry effects the discussion will focus on the regression without industry controls, although the discussion can be readily applied for the industry controlled regressions.
The relationship between firm size and leverage varies across the remaining three capital structure and financing variables.
From Table 3 a ten-fold increase in the size of the firm corresponds to a 3. A positive relationship is also found for long-term leverage, con- sistent with information asymmetries and transaction cost arguments limiting the attractiveness of debt, in particular long-term debt. This suggests that the nature of financing, such as duration, rather than the source of financing may be causing the underlying affect upon capital structure.
Examining NONA, an extra 10 per cent of fixed assets leads to a reduction of leverage of 1. This is the opposite predicted by finance theory. An explanation for this result can be found by examining the effect of NONA on long and short-term leverage. In particular, for an extra 10 per cent of fixed assets, long-term leverage is increased by 2. The inverse relationship between non-current assets and short-term leverage is consistent with firms matching their duration of assets and liabilities.
Given the larger proportion of short-term debt over long-term debt most likely explains why the coefficient for NONA on firm leverage is negative. The relationship between NONA and bank financing is significantly positive, as predicted, however, the relationship between NONA and outside financing is actually negative.
This discrepancy between these two financing variables suggests that banks place more weight upon the fixed assets of the firm than other financiers, and that these other outside financiers do not tend to rely on fixed assets to reduce agency costs. Consistent with pecking order arguments, in all five regressions, the coefficient for ROA is negative and significant. The strongest effect appears to be for leverage, while the smallest effect of the five variables is for bank financ- ing.
GROW is positive for all five dependent variables and significant for leverage, short-term leverage and out- side financing. Interestingly, bank finance, while positive, was not significant suggesting that firms that grow appear to use other sources of outside financing to support their growth. The analyses were replicated using the one-year lagged variables instead of the three-year average, with the exception of RISK which was estimated again over a three-year period.
The results from this ana- lysis were very similar to the three-year average results. Given these counter-intuitive findings, this raises the question as to whether risk is im- portant in the capital structure of SMEs, or whether our risk measure is capturing this construct.
To investigate the second issue, we re-ran the regressions in Table 3 with an alternative risk measure ALTRISK applied previously by Wald in examining relatively larger firms.
The results of these regressions are displayed in Appendix B. Interestingly, the coefficients on the alternative risk measure are all negative, with LEV, LONG and BANK all exhibiting reasonably strong effects, consistent with the arguments that high risk firms should find debt less attractive.
In addition, all the other variables exhibit similar relationships to the regressions using the original RISK measure. Overall the tobit specification provided similar inferences across the various variables and financing choices. The one exception being the negative coefficients between firm size and all the financing variables except long-term financing. Therefore the results as discussed, in particular between long-term financing and size are robust to the tobit specification, with the other variables which were insignificant becoming on occasion significantly negative.
This suggests that aside from long-term leverage, scale effects appear to be a limited influence upon the capital structure of SMEs. Interaction between size and financing While this study has focused on a selection of firms within the small and medium sized classification, there is a possibility that the relationships between the financing and firm characteristics do vary by size within the sample.
In particular, the same influences that may cause differences between SMEs and larger listed firms, may also affect relationships within the SME group, due to the wide variation of sizes present.
To empirically examine the presence of size interaction effects, the sample was partitioned into two groups, above the sample median for total assets, and below the sample median for total assets. The results can be interpreted as if two separate studies were undertaken with different size criteria employed, with control of overall industry effects.
The regressions with interaction effects between firm size and the dependent variables are reported in Table 4. SIZE appears to exhibit a similar effect across size groups, with leverage positively related to size. The relationship between NONA and all financing variables appears to be consistent across size groups for all variables. In addition, GROW appears to be fairly con- sistent across size groups, with the exception being the non-significant negative coefficient on bank financing.
Finally, RISK is only significant for BIG firms in regard to the leverage variables, again in the opposite direction to that hypothesised. Interestingly, the relationship between RISK and the propor- tion of bank financing for the SMALL firms in the sample suggests that higher risk reduces the use of bank financing. Therefore, aside from size, the influence of the variables upon the capital structure choice appear to be homogenous across the sample. This suggests that the extreme firms of the sample, being small or large, are not driving the inferences found.
Implications and limitations The clear implication from this study is the relevance of financial theories of capital structure and financing applying to Australian SMEs. The size interaction results show that the asset structure, profitability and growth findings are fairly robust across size groups within the SME classification.
This study has also highlighted the importance of distinguishing between long and short forms of debt when making inferences about capital structure choice. In particular, the role asset structure plays within the capital structure of the firm. Given the relatively high proportion of short term debt financing in these firms, overall leverage is negatively related to fixed assets. However, dividing the duration of debt into long and short components, it is found that long-term debt structure is positively related to long-term asset structure.
This is intuitive both from a theoretical perspective and from a duration matching perspective. Another interesting implication from this study is how bank financing differs from other outside financing.
0コメント