Do country or firm factors explain capital structure? Evidence from SMEs in France and Greece

Abstract

We investigate the capital structure determinants of small and medium sized enterprises (SMEs) using a sample of Greek and French firms. We address the following questions: Are the capital structure determinants of SMEs in the two countries driven by similar factors? Are potential differences driven by country-specific or firm-specific factors? Are the size and structure of their financial markets important factors to explain any cross-country differences on SME capital structure? To answer these questions we apply panel data methods to the sample of firms for the period 1998 to 2002. We assess the extent to which the debt to assets ratio of firms depends upon their asset structure, size, profitability and growth rate. The results show that the SMEs in both countries exhibit similarities in their capital structure choices. Asset structure and profitability have a negative relationship with leverage, whereas firm size is positively related to their debt to assets ratio. Growth is statistically significant only for France and is positively related to debt. We attribute these similarities to their institutional characteristics and in particular the commonality of their civil law systems. We find differences in the intensity of the capital structure relationship between the two countries. We provide evidence that these differences are due to firm-specific rather than country factors.

Acknowledgements

We would like to thank Dimitris Margaritis, an anonymous referee and the Editor for their helpful comments.

Notes

¹ See the 7th Report on SMEs of the European Commission (2002).

² For example, see Bradley et al. (1984), Chen and et al. 1998), Wald (1999).

³ Recommendation 96/280/EC, 3 April 1996.

⁴ On 6 May 2003 the Commission adopted a new Recommendation 2003/361/EC regarding the SME definition which replaced Recommendation 96/280/EC as from 1 January 2005 (see Commission of the EC 2003). According to the new definition SMEs must have an annual turnover of <50 million euros, and an annual balance sheet of <43 million euros.

⁵ ATHEX (2003), Fact Book, Department of Information Dissemination Marketing Division, p. 19.

⁶ For a review of the literature on capital structure, see Harris and Raviv (1991) and Myers 2001.

⁷ For instance, Rajan and Zingales (1995) provide an extensive analysis of the determinants of capital structure by examining the financing decisions of public firms in the major industrialized countries.

⁸ Note that tax advantages of borrowed money are assumed second-order in the pecking order model.

⁹ In fact, Vasiliou and Daskalakis (2005) provide an extensive analysis of the capital structure for the listed Greek firms and find that these firms avoid long-term debt financing and turn to short-term debt. They imply that this may be due to the bank policy in Greece and if we consider that this implication does hold for the listed firms, there seems to be no reason why this implication should not also hold for the SMEs.

¹⁰ The model is estimated, for the period 1998 to 2002 because we allow for one year lag in our empirical analysis.

¹¹ We measure leverage considering the ratio of total liabilities to total assets. Total liabilities include leasing, accounts payable and accounts receivable, namely the trade credit which is an important mean of finance of SMEs'. For this reason we consider this broader definition.

¹² For instance in 2002 according to the same source only about 60% of the SMEs regularly provide documents such as the balance sheet and the profit and loss statement. Around 8% of the SMEs hand over to their financier their annual budget, whilst 7% also share financial plans or cash flows forecasts with them and about 4% provide information on inventories or unpaid invoices.

¹³ We also thought of considering the ‘age’ of the firms as a capital structure determinant in our analysis. The age of the firm is another variable that is featured in cross-country capital structure studies. We decided not to include this variable for two reasons: (1) it will effectively amount to introducing a trend in a balanced panel; and (2) the mortality rate for SMEs is higher than 50% during the first five years of their operation.

¹⁴ However, note that one possible disadvantage is that variables such as age cannot be included as regressors.

¹⁵ For more information regarding panel modes characterstics see Chamberlain (1998), Hausman and Taylor (1981) and Judge et al. (1985).

¹⁶ We do not control for industry specific effects as differences in capital structure are likely to be due to firm specific factors rather than industry differences (Myers, 1984).

¹⁷ We have also estimated the panel regression model using lagged values of the asset structure, profitability and growth variables. The results are very similar suggesting that it is unlikely that the reported estimates are subject to serious simultaneity bias problems.