- Email: [email protected]
Financial liberalisation and Capital structuring decisions of corporate firms: Evidence from India
Economics Letters 149 (2016) 33–37
Contents lists available at ScienceDirect
Economics Letters journal homepage: www.elsevier.com/locate/ecolet
Financial liberalisation and Capital structuring decisions of corporate firms: Evidence from India Nemiraja Jadiyappa ∗ , Nagi Reddy Vanga, Raveesh Krishnankutty IFHE (IBS Hyderabad), India
highlights • • • • •
The impact financial liberalisation on leverage and debt maturity is negative. Debt specialisation increased following the liberalisation. The effect is more pronounced for priority firms compared to non priority firms. We use a reform index to measure financial liberalisation. Reforms index captures the gradual nature of implementation of financial reforms.
article
info
Article history: Received 27 August 2016 Received in revised form 3 October 2016 Accepted 7 October 2016 Available online 17 October 2016
abstract We examine the impact of financial reforms on corporate financing decisions of Indian firms using the fixed effects panel estimator. The impact on the leverage and debt maturity ratio is negative while a positive effect is observed on debt specialisation. The impact is greater for priority firms compared to non-priority firms. © 2016 Elsevier B.V. All rights reserved.
JEL classification: G0 G3 G15 G20 G30 Keywords: Capital structure Financial reforms Debt maturity structure Debt specialisation Financial liberalisation
1. Introduction At some time in history, most countries followed regulated financial systems which included one or more of the following forms: administered interest rates, directed credit schemes, subsidised loans, capital controls and credit ceiling. Verma (1998) observed that a ‘‘financial repression state1 ’’ affects the efficient allocation of resources and interferes with lending and investment decision making which results in distorted credit allo-
∗ Correspondence to: IBS Hyderabad, A Constituent of IFHE, Deemed to be University, Hyderabad, Telangana, India-501203. E-mail address: [email protected] (N. Jadiyappa). 1 Heavily regulated financial systems. http://dx.doi.org/10.1016/j.econlet.2016.10.004 0165-1765/© 2016 Elsevier B.V. All rights reserved.
cation (Blundell-Wignall and Browne, 1991). These distortions/ imperfections at the macro level would then affect corporate financing decisions at the organisational (firm) level. The adopted financial reforms generally aim at correcting these distortions by influencing structural aspects such as size, efficiency and competitiveness of financial markets, which in turn determine the firm’s policies on (a) capital structure (sources of capital); (b) debt maturity structure (type of debt capital), and (c) debt specialisation (number of sources of debt capital) that they want to use. It is therefore important to empirically examine how financial reforms change corporate financing decisions at the firm level. This kind of empirical examination assumes greater importance in the context of financial reforms being undertaken by many developing countries. The initiation of Indian financial liberalisation in the early 1990s provides the best setting for the examination of these
34
N. Jadiyappa et al. / Economics Letters 149 (2016) 33–37
issues. Hence, this study examines the impact of financial reforms on the corporate financing decisions of Indian firms in three areas: (a) Capital structure—decisions in this area determine the different proportions of debt and equity in the overall capital. (b) Debt maturity structure—for a given level of debt, decisions in this area deal with the choice of debt instruments based on their maturity (long term vs. short term). (c) Debt specialisation—decisions related to this topic determine the number of debt sources a firm would want to use for a given level of debt and debt maturity structure, i.e. the extent of diversification/specialisation in firms with respect to their debt sources. The following section describes four important changes that have been taken place in the Indian financial system following the initiation of financial liberalisation in 1991. Assuming that these macro changes will influence the micro behaviour of firms with respect to their financing decisions, these changes have been used as a base to develop the present study’s hypotheses in Section 2. 1.1. Structural changes in the Indian financial system following reforms The following structural changes are observed as a result of financial reforms: (a) Change in the relative dominance of stock markets over credit markets (Fig. 1). Stock markets became a more important source of capital for firms following the introduction and implementation of reforms. (b) Development of the Indian banking system in terms of its size, efficiency, competition and geographical spread (Jadiyappa and Mitra, 2012). (c) Marked reduction (if not abolishment) of governmental intervention in the financial market through interest subsidy schemes, credit allocation, credit ceilings and so on since the inception of financial reforms (Reddy, 2002; Khanna, 1999). (d) Increase in the rate of financial innovations. Financial reforms increase the rate of financial innovations (Blundell-Wignall and Browne, 1991) and therefore firms have greater freedom for choosing debt instruments which meet their requirements thereby resulting in capital structure and debt maturity changes. 2. Hypotheses 2.1. Financial reforms and capital structure All the above mentioned four structural changes in the financial system should influence capital structure decisions at the firm level. The first and third arguments predict a negative impact while the second one predicts a positive impact on the debt level. The impact of financial innovations on debt policy is not very clear. The results of an aggregate level analysis would clearly show the outcome of the interaction between these forces. For example, large firms are expected to benefit from stock market development, whereas smaller firms are expected to benefit from banking development. However, the extent of benefit in both cases depends on how they were regulated financially in the pre-reforms period.2 Consequently, in the present study a sub-hypothesis was formulated which examines the impact of reforms on two categories of firms, namely priority firms (those which were favourably regulated), and non-priority firms (those which were unfavourably regulated) in the pre-reforms period.
2 Priority firms vs.\non-priority firms.
Fig. 1. Movement of the ratio of stock market capitalisation/GDP to domestic credit to the private sector by banks/GDP in India during 1988–2005. Source: World Bank.
With the abolition of government sponsored credit schemes, priority firms will no longer get credit at low rates and this would cause their debt level to decline. Moreover, the development of banking systems removes many inherent distortions in the system, and this induces priority firms to decrease their leverage ratio.3 For non-priority firms, the liberalisation process opens up the financial system and affords them an opportunity to raise debt capital at relatively attractive rates. Furthermore, for this subset of firms, leverage is expected to increase. The net effect of these changes at the country level would then depend on the proportion of firms falling under the priority and non-priority sectors. Thus, the following two hypotheses were formulated: H1: Financial reforms affect the leverage ratio of firms. H1A: The impact of financial reforms on the leverage ratio of priority firms (non-priority firms) is negative (positive) 2.2. Financial liberalisation, debt structure and specialisation analysis This section evaluates how firms existed under financially regulated system changed their debt maturity patterns once that financial system was liberalised. Demirgüç-Kunt and Maksimovic (1999) note that government credit subsidy schemes are likely to favour long-term debt, and this would result in a higher proportion of long term debt for a given level of debt. With the abolition of these schemes however, this ratio should gradually decline at least for priority firms. Moreover, both the agency theory4 as well as the signalling theory5 also predict a negative impact of financial reforms on the proportion of long term debt and a resultant decline in the proportion of long term debt even for non-priority firms. Thus Hypothesis 2 is as follows: H2: Financial reforms impact negatively the long term debt proportion of firms. There is a very minimal level of financial innovation6 in the regulated financial systems and any reforms in these systems would increase the rate of financial innovations (Blundell-Wignall and Browne, 1991). In the absence of diversified financial instruments, firms have no other option but to use the available
3 As they were over levered in the pre-reform period, which increases the risk of bankruptcy thus lowering the chances of obtaining a new loan. 4 The agency theory predicts that in a liberalised system, with more discretion at their disposal, banks have an incentive to issue more short term debt as this allows them to monitor managers’ actions more effectively. 5 The signalling framework hypothesises that in the presence of information asymmetry, high value/quality firm would issue short term debt as the sensitivity to mispricing is lower for short term debt and financial liberalisation is expected to develop debt markets (bond market) where the question of information asymmetry plays an important role. 6 Measured in terms of number of financial instruments available for firms.
N. Jadiyappa et al. / Economics Letters 149 (2016) 33–37
instruments irrespective of their suitability for meeting their requirements. Thus, a high degree of debt specialisation (i.e., the use of one dominant type of debt) is expected in such regulated systems. The effect of financial reforms on debt specialisation patterns is ambiguous. Availability of more financial instruments favours the view that debt specialisation declines following liberalisation, whereas on the other hand, patterns observed in developed countries (those with developed financial markets) suggest that it would increase.7 Thus, the following hypothesis was formulated to test for debt specialisation changes following financial reforms. H3: Financial reforms change the debt specialisation pattern at the firm le vel.
35
Table 1 Impact of financial reforms on corporate financing decisions. Variables
Model I Leverage
Model II Leverage
Model III LTM
Model IV DSR
Intercept
1.398 (0.027) −0.065 (0.021) −0.241 (0.012) 0.084 (0.029) −0.355 (0.028) −0.044 (0.005)
1.40 (0.027) −0.103 (0.028) −0.241 (0.012) 0.08 (0.029) −0.355 (0.028) −0.044 (0.005) 0.074 (0.033)
0.302 (0.016) −0.029 (0.01) −0.004 (0.005) 0.269 (0.015) −0.071 (0.011) 0.015 (0.003)
0.392 (0.013) 0.089 (0.009) −0.018 (0.005) 0.017 (0.012) −0.116 (0.012) 0.003 (0.002)
Leveraget −1
−0.052
−0.152
R-square F No. of observations No. of firms
(0.006) 0.048 78.38 33 873 2834
(0.005) 0.099 170.98 33 079 2834
RI Size Tangibility ROAt −1 Growth PRRI
3. Methodology and data 3.1. Variables and their measurement 3.1.1. Dependent variables Leverage ratio: The ratio of total debt to total assets. Long term debt maturity ratio (LTM): the ratio of long term debt to total debt is the long term debt maturity ratio (LTM). By following the Barclay and Smith approach8 (Barclay and Smith, 1995) debts with a maturity period of more than one year were considered as long term debts. Debt Specialisation Ratios (DSR): DSR is defined as the sum of the squared proportion of individual type of debts9 that constitute the total debt (Colla et al., 2011). 3.1.2. Independent and control variables The main independent variable in this study was the index of financial liberalisation adopted from Abiad et al. (2008). These researchers developed a database of financial reforms for many economies/countries to capture the gradual and stepwise developments in the following seven broad dimensions of regulation in the financial market: credit control and reserve requirements; external finance control; interest rate controls; bank entry barriers; bank privatisation; securities markets and bank supervision. Abiad et al. (2008) assigned a liberalisation score for each of the seven dimensions based on the extent of liberalisation, and then summed up the individual scores to arrive at the final score. Over the years, the index value changes based on the liberalisation measures adopted in any particular dimension. Detailed description of the index/database can be found in Abiad et al. (2008). As long as the liberalisation measure has sufficient time variance (and this variation is significant enough to prompt a response), it should be possible to conduct effective tests of the hypotheses laid out in the preceding section. 3.2. Model specification The base model used for estimation is presented in Eq. (1) Yjt = αj + β1 RI t + β2 Sizejt + β3 Tangibilityjt
+ β4 ROAjt −1 + β5 Growthjt + εj
(1)
7 Firms in developed financial markets have high degree of debt specialisation because firms choose that type of debt which suits its need the best (Rauh and Sufi, 2010). 8 Barclay and Smith (1995) consider debt with remaining maturity period of more than three years as the long term debt. But, they tested their hypotheses using debt with more than one year to maturity also. 9 We consider eight types of debt. They are bank debt, debt from financial institutions, government debt, loans from promoters, inter-corporate loans, deferred credit, trade credit and foreign currency borrowings.
0.142 292.46 33 873 2834
0.143 245.67 33 873 2834
Dependent variables: Leverage is the ratio of total debt to total assets. LTM is the ratio of long term debt to total debt, DSR is defined as the sum of the squared ratio of individual type of debt over total debt. Independent variables: RI is the financial reforms index adopted from Abiad et al. (2008), size is log (sales), tangibility is the ratio of net fixed assets to total assets, ROA is return on assets, growth is the annual growth rate in total assets. PRRI is an interaction dummy which is the product of RI and Priority dummy (Priority firms take the value zero and non-priority firms one). The parameters are estimated using the fixed effects estimation procedure. In parenthesis are the robust standard errors clustered at firm level.
where, Size: Following the methodology of Rajan and Zingales (1995) we measure size as the log of yearly sales. Growth: The annual growth rate in total assets is a growth proxy that has been used earlier by Fama and French (2001). Tangibility: The ratio of net fixed assets to total assets was used as a measure of tangibility. Profitability; To proxy for performance (Profitability) we use Return on Assets (ROA) which is defined as the ratio of profit before tax, interest and depreciation to total assets. Lagged values of ROA were used because size and ROA variables are endogenously related.10 The parameters were estimated using the fixed effects estimator.11 3.3. Data Firm level data was taken from the database of Centre for Monitoring Indian Economy (CMIE) which contains information of listed and unlisted firms from 1988 onwards. For the purpose of analysis, the present study used firm data which spans the period 1988 through 2005.12 The end point for data was dictated by the measure for liberalisation, viz. reforms index, which ended in 2005. Firms were chosen based on two criteria: (i) they should have existed before 1992, and (ii) they should have at least three years of leverage data for the period 1988–1992.13
10 Huselid et al. (1997) argue that firm size is one of the important determinants of firm performance. 11 Haushman test rejected the null at 1%. 12 Though our study period begins in 1988, the actual financial liberalisation took place, according to the International Monetary Fund in November, 1991. Thus, index value from 1988 to 1991 is very minimal. 13 This is required to classify firms into priority and non-priority firms.
36
N. Jadiyappa et al. / Economics Letters 149 (2016) 33–37 Table 2 Correlation matrix. Variables
Leverage
RI
Size
Leverage RI Size Tangibility ROAt −1 Growth
1.000 −0.125 −0.209 0.013 −0.256 −0.125
1.000 0.073 0.041 −0.150 −0.082
−0.125
Tangibility
ROAt −1
Growth
1.000 −0.084 −0.088
1.000 0.164
1.000
1.000 0.198 0.144
Leverage is the ratio of total debt to total assets. RI is the reforms index adopted from Abiad et al. (2008), size is log (sales), tangibility is the ratio of net fixed assets to total assets, ROA is return on assets, AGR is the annual growth rate in total assets.
4. Results and discussion 4.1. Hypothesis H1 and H1A In the first hypothesis, it was argued that financial reforms will have a significant impact on the leverage ratio of firms because of the reasons mentioned in the previous sections. The results are presented in Table 1. As hypothesised, the RI coefficient, which represents the reforms, is negative and significant, implying a negative impact of financial reforms on the debt level of firms. This result supports the findings of Bertrand et al. (2007) and Senay et al. (2007) who respectively observed the same results in the French and the developing world context. Bertrand et al. (2007) attribute this decrease to the abolition of government credit schemes in the post-reform period while Senay et al. (2007) ascribed it to the reforms which were aimed at minimising the impact of imperfections in the legal, capital and administrative spheres. In India as observed by Reddy (2002), the observed decline in debt ratios can be attributed to both the causes. These two reasons are complementary to each other and not mutually exclusive i.e., government loans would stop coming because of reforms and firms will switch to stock markets to fund their new growth.14 The next step involved analysis of the differential impact of financial liberalisation on priority and non-priority firms.15 Following the approach of Bertrand et al. (2007) firms were classified into priority and non-priority groups based on the median value of average long term leverage for the period 1988–1992. The results are presented in the third column of Table 1. The main variable of interest in this analysis was the product term (PRRI) which interacts the effects of financial reforms (RI) with the priority and nonpriority classification (PR). This interaction term gives a differential slope coefficient for non-priority firms.16 As is evident, the RI variable is negative and significant, indicating a negative impact on priority firms which is consistent with the study’s prediction. The interaction coefficient was found to be positive, but the adjusted slope coefficient was negative for non-priority firms as well. This is contrary to what had been predicted in H1A. However, it should be noted that the rate of decrease is significantly lower for nonpriority firms. 4.2. Hypothesis H2 Debt maturity and debt specialisation analysis As discussed earlier, financial liberalisation should exert a negative effect on LTM ratio. The results are presented in the fourth
14 As bank dominated financial system progress towards stock market dominated system, it is expected that firms start substituting debt by external equity to fund their investments. Consequently, debt level would decline. 15 Financial regulation in the pre-reforms period consisted of a complex web of rules and regulations. Any credible classification system should consider this complex web of regulations. In absence of information on such a complex system, we adopt an indirect method to classify firms into priority and non-priority firms as suggested by Bertrand et al. (2007). 16 Non priority firms take value one in PR dummy.
column of Table 1 and provide evidence to support the Hypothesis H2, because RI variable is negative and significant (−0.029). If there is truth in the argument that the negative impact is because of abolition of government support, then the observed negative impact in column four should be driven by the negative impact on the priority firms. This supposition was verified in an unreported analysis by including PRRI variable (interaction variable between priority dummy and RI) into Model III of Table 1. The results were found to be exactly as predicted with priority firms having a negative coefficient and non-priority firms having a positive adjusted coefficient. The findings in the present study contradict the result of Senay et al. (2007) who found no significant impact of financial reforms on the debt maturity ratio of firms in emerging countries. This might be due to differences in the starting points of financial reforms in various countries included in their sample, which is a very common problem in a cross-country study like theirs. 4.3. Hypothesis H3 Lastly, the results of the debt specialisation analysis are presented in the last column of Table 1. The debt specialisation ratio (DSR) that was used in our analysis is negatively related to debt diversification. That is to say, higher values of DSR indicate higher debt concentration; in other words, firms use a few dominant sources of debt. It is very clear that financial liberalisation caused an increase in the concentration of debt and a decrease in the diversification of debt sources as the RI co-efficient is positive and significant. On further analysis of this observation, it was found that the observed decrease in debt diversification is mainly due to substitution of informal sources of debt17 by the formal sources of debt like bank debt. Therefore, it can be stated that in the Indian context, financial reforms increased the access to formal sources of debt. 5. Conclusion This study shows how financial reforms could affect the behaviour of firms at the micro level. This result assumes great importance in the context of the introduction and implementation of financial reforms in many developing countries. The likely behavioural changes that these reforms would induce at the firm level are of great significance to both policy makers and investors alike. Since reform measures are converging across countries along the lines of developed world practices, the results from studies on individual countries could easily be generalisable across other countries too. It is very likely that the results of this study namely, a decline in debt/debt maturity and an increase in specialisation levels, would also be manifested in other countries which initiate reforms in their financial systems. Appendix A See Table 2.
17 Like inter-corporate loans, loans from promoters, friends, relatives,etc.
N. Jadiyappa et al. / Economics Letters 149 (2016) 33–37 Table 3 Regression results after clustering the standard errors at industry level.
References
Variables
Leverage Model I
Leverage Model II
LTM Model III
DSR Model IV
Intercept
1.171 (0.028) −0.266 (0.028) −0.093 (0.011) −0.048 (0.027) −0.622 (0.049) −0.064 (0.006)
1.183 (0.028) −0.113 (0.028) −0.094 (0.011) −0.090 (0.026) −0.627 (0.048) −0.064 (0.006) −0.282 (0.023)
0.139 (0.016) −0.072 (0.015) 0.018 (0.005) 0.465 (0.019) −0.031 (0.014) 0.030 (0.003)
0.357 (0.013) 0.088 (0.011) −0.024 (0.003) 0.016 (0.011) −0.102 (0.015) 0.002 (0.002)
0.011 (0.008) 0.151 102.720 33 873 2834
−0.086
RI Size Tangibility ROAt −1 Growth PRRI Leveraget −1 R-square F No. of observations No. of firms
0.123 180.790 33 873 2834
0.150 171.230 33 873 2834
(0.005) 0.054 81.370 33 079 2834
Dependent variables: Leverage is the ratio of total debt to total assets. LTM is the ratio of long term debt to total debt, DSR is defined as the sum of the squared ratio of individual type of debt over total debt. Independent variables: RI is the financial reforms index adopted from Abiad et al. (2008), size is log (sales), tangibility is the ratio of net fixed assets to total assets, ROA is return on assets, growth is the annual growth rate in total assets. PRRI is an interaction dummy which is the product of RI and Priority dummy (Priority firms take the value zero and non-priority firms one). The parameters are estimated using the fixed effects estimation procedure. In parenthesis are the robust standard errors clustered at industry level.
Appendix B See Table 3.
37
Abiad, Abdul, Tressel, Thierry, Detragiache, Enrica, 2008. A New Database of Financial Reforms. IMF Working Papers (08/266). Barclay, Michael J., Smith Jr., Clifford W., 1995. The maturity structure of corporate debt. J. Finance 50 (2), 609–631. Bertrand, Marianne, Schoar, Antoinette, Thesmar, David, 2007. Banking deregulation and industry structure: Evidence from the French banking reforms of 1985. J. Finance 62 (2), 597–628. Blundell-Wignall, A., Browne, F., 1991. Macroeconomic Consequences of Financial Liberalisation: A Summary Report, OECD Economics Department Working Papers, No. 98, OECD Publishing. http://dx.doi.org/10.1787/088742707378. Colla, Paolo, Ippolito, Filippo, Li, Kai, 2011. Debt Specialization. SSRN eLibrary. http://dx.doi.org/10.2139/ssrn.1520902. Demirgüç-Kunt, Asli, Maksimovic, Vojislav, 1999. Institutions, financial markets, and firm debt maturity. J. Financ. Econ. 54 (3), 295–336. http://dx.doi.org/10. 1016/s0304-405x(99)00039-2. Fama, Eugene F., French, Kenneth R., 2001. Disappearing dividends: Changing firm characteristics or lower propensity to pay? J. Appl. Corp. Financ. 14 (1), 67–79. http://dx.doi.org/10.1111/j.1745-6622.2001.tb00321.x. Huselid, M.A., Jackson, S.E., Schuler, R.S., 1997. Technical and strategic human resources management effectiveness as determinants of firm performance. Acad. Manag. J. 40 (1), 171–188. Jadiyappa, Nemiraja, Mitra, Amarnath, 2012. Financial Regulations, Structure of Financial Systems and Capital Structure, August 30, 2012. Available at SSRN: http://ssrn.com/abstract=2138660 or http://dx.doi.org/10.2139/ssrn.2138660. Khanna, Sushil, 1999. Financial reforms and industrial sector in India. Econ. Polit. Weekly 34 (45), 3231–3241. Rajan, Raghuram G., Zingales, Luigi, 1995. What do we know about capital structure? Some evidence from international data. J. Finance 50 (5), 1421–1460. no.. Rauh, Joshua D., Sufi, Amir, 2010. Capital structure and debt structure. Rev. Financ. Stud. 23 (12), 4242–4280. http://dx.doi.org/10.1093/rfs/hhq095. Reddy, Venugopal, 2002. Monetary and Financial Sector Reforms in India: A Practitioner’s Perspective. Senay, Agca, NicolÃ, Gianni De, Detragiache, Enrica, 2007. Financial Reforms, Financial Openness, and Corporate Borrowing: International Evidence. IMF Working Papers (07/186). Verma, Jayanth, 1998. Indian financial sector reforms: A corporate perspective. Vikalpa 23 (1), 27–38.
Contents lists available at ScienceDirect
Economics Letters journal homepage: www.elsevier.com/locate/ecolet
Financial liberalisation and Capital structuring decisions of corporate firms: Evidence from India Nemiraja Jadiyappa ∗ , Nagi Reddy Vanga, Raveesh Krishnankutty IFHE (IBS Hyderabad), India
highlights • • • • •
The impact financial liberalisation on leverage and debt maturity is negative. Debt specialisation increased following the liberalisation. The effect is more pronounced for priority firms compared to non priority firms. We use a reform index to measure financial liberalisation. Reforms index captures the gradual nature of implementation of financial reforms.
article
info
Article history: Received 27 August 2016 Received in revised form 3 October 2016 Accepted 7 October 2016 Available online 17 October 2016
abstract We examine the impact of financial reforms on corporate financing decisions of Indian firms using the fixed effects panel estimator. The impact on the leverage and debt maturity ratio is negative while a positive effect is observed on debt specialisation. The impact is greater for priority firms compared to non-priority firms. © 2016 Elsevier B.V. All rights reserved.
JEL classification: G0 G3 G15 G20 G30 Keywords: Capital structure Financial reforms Debt maturity structure Debt specialisation Financial liberalisation
1. Introduction At some time in history, most countries followed regulated financial systems which included one or more of the following forms: administered interest rates, directed credit schemes, subsidised loans, capital controls and credit ceiling. Verma (1998) observed that a ‘‘financial repression state1 ’’ affects the efficient allocation of resources and interferes with lending and investment decision making which results in distorted credit allo-
∗ Correspondence to: IBS Hyderabad, A Constituent of IFHE, Deemed to be University, Hyderabad, Telangana, India-501203. E-mail address: [email protected] (N. Jadiyappa). 1 Heavily regulated financial systems. http://dx.doi.org/10.1016/j.econlet.2016.10.004 0165-1765/© 2016 Elsevier B.V. All rights reserved.
cation (Blundell-Wignall and Browne, 1991). These distortions/ imperfections at the macro level would then affect corporate financing decisions at the organisational (firm) level. The adopted financial reforms generally aim at correcting these distortions by influencing structural aspects such as size, efficiency and competitiveness of financial markets, which in turn determine the firm’s policies on (a) capital structure (sources of capital); (b) debt maturity structure (type of debt capital), and (c) debt specialisation (number of sources of debt capital) that they want to use. It is therefore important to empirically examine how financial reforms change corporate financing decisions at the firm level. This kind of empirical examination assumes greater importance in the context of financial reforms being undertaken by many developing countries. The initiation of Indian financial liberalisation in the early 1990s provides the best setting for the examination of these
34
N. Jadiyappa et al. / Economics Letters 149 (2016) 33–37
issues. Hence, this study examines the impact of financial reforms on the corporate financing decisions of Indian firms in three areas: (a) Capital structure—decisions in this area determine the different proportions of debt and equity in the overall capital. (b) Debt maturity structure—for a given level of debt, decisions in this area deal with the choice of debt instruments based on their maturity (long term vs. short term). (c) Debt specialisation—decisions related to this topic determine the number of debt sources a firm would want to use for a given level of debt and debt maturity structure, i.e. the extent of diversification/specialisation in firms with respect to their debt sources. The following section describes four important changes that have been taken place in the Indian financial system following the initiation of financial liberalisation in 1991. Assuming that these macro changes will influence the micro behaviour of firms with respect to their financing decisions, these changes have been used as a base to develop the present study’s hypotheses in Section 2. 1.1. Structural changes in the Indian financial system following reforms The following structural changes are observed as a result of financial reforms: (a) Change in the relative dominance of stock markets over credit markets (Fig. 1). Stock markets became a more important source of capital for firms following the introduction and implementation of reforms. (b) Development of the Indian banking system in terms of its size, efficiency, competition and geographical spread (Jadiyappa and Mitra, 2012). (c) Marked reduction (if not abolishment) of governmental intervention in the financial market through interest subsidy schemes, credit allocation, credit ceilings and so on since the inception of financial reforms (Reddy, 2002; Khanna, 1999). (d) Increase in the rate of financial innovations. Financial reforms increase the rate of financial innovations (Blundell-Wignall and Browne, 1991) and therefore firms have greater freedom for choosing debt instruments which meet their requirements thereby resulting in capital structure and debt maturity changes. 2. Hypotheses 2.1. Financial reforms and capital structure All the above mentioned four structural changes in the financial system should influence capital structure decisions at the firm level. The first and third arguments predict a negative impact while the second one predicts a positive impact on the debt level. The impact of financial innovations on debt policy is not very clear. The results of an aggregate level analysis would clearly show the outcome of the interaction between these forces. For example, large firms are expected to benefit from stock market development, whereas smaller firms are expected to benefit from banking development. However, the extent of benefit in both cases depends on how they were regulated financially in the pre-reforms period.2 Consequently, in the present study a sub-hypothesis was formulated which examines the impact of reforms on two categories of firms, namely priority firms (those which were favourably regulated), and non-priority firms (those which were unfavourably regulated) in the pre-reforms period.
2 Priority firms vs.\non-priority firms.
Fig. 1. Movement of the ratio of stock market capitalisation/GDP to domestic credit to the private sector by banks/GDP in India during 1988–2005. Source: World Bank.
With the abolition of government sponsored credit schemes, priority firms will no longer get credit at low rates and this would cause their debt level to decline. Moreover, the development of banking systems removes many inherent distortions in the system, and this induces priority firms to decrease their leverage ratio.3 For non-priority firms, the liberalisation process opens up the financial system and affords them an opportunity to raise debt capital at relatively attractive rates. Furthermore, for this subset of firms, leverage is expected to increase. The net effect of these changes at the country level would then depend on the proportion of firms falling under the priority and non-priority sectors. Thus, the following two hypotheses were formulated: H1: Financial reforms affect the leverage ratio of firms. H1A: The impact of financial reforms on the leverage ratio of priority firms (non-priority firms) is negative (positive) 2.2. Financial liberalisation, debt structure and specialisation analysis This section evaluates how firms existed under financially regulated system changed their debt maturity patterns once that financial system was liberalised. Demirgüç-Kunt and Maksimovic (1999) note that government credit subsidy schemes are likely to favour long-term debt, and this would result in a higher proportion of long term debt for a given level of debt. With the abolition of these schemes however, this ratio should gradually decline at least for priority firms. Moreover, both the agency theory4 as well as the signalling theory5 also predict a negative impact of financial reforms on the proportion of long term debt and a resultant decline in the proportion of long term debt even for non-priority firms. Thus Hypothesis 2 is as follows: H2: Financial reforms impact negatively the long term debt proportion of firms. There is a very minimal level of financial innovation6 in the regulated financial systems and any reforms in these systems would increase the rate of financial innovations (Blundell-Wignall and Browne, 1991). In the absence of diversified financial instruments, firms have no other option but to use the available
3 As they were over levered in the pre-reform period, which increases the risk of bankruptcy thus lowering the chances of obtaining a new loan. 4 The agency theory predicts that in a liberalised system, with more discretion at their disposal, banks have an incentive to issue more short term debt as this allows them to monitor managers’ actions more effectively. 5 The signalling framework hypothesises that in the presence of information asymmetry, high value/quality firm would issue short term debt as the sensitivity to mispricing is lower for short term debt and financial liberalisation is expected to develop debt markets (bond market) where the question of information asymmetry plays an important role. 6 Measured in terms of number of financial instruments available for firms.
N. Jadiyappa et al. / Economics Letters 149 (2016) 33–37
instruments irrespective of their suitability for meeting their requirements. Thus, a high degree of debt specialisation (i.e., the use of one dominant type of debt) is expected in such regulated systems. The effect of financial reforms on debt specialisation patterns is ambiguous. Availability of more financial instruments favours the view that debt specialisation declines following liberalisation, whereas on the other hand, patterns observed in developed countries (those with developed financial markets) suggest that it would increase.7 Thus, the following hypothesis was formulated to test for debt specialisation changes following financial reforms. H3: Financial reforms change the debt specialisation pattern at the firm le vel.
35
Table 1 Impact of financial reforms on corporate financing decisions. Variables
Model I Leverage
Model II Leverage
Model III LTM
Model IV DSR
Intercept
1.398 (0.027) −0.065 (0.021) −0.241 (0.012) 0.084 (0.029) −0.355 (0.028) −0.044 (0.005)
1.40 (0.027) −0.103 (0.028) −0.241 (0.012) 0.08 (0.029) −0.355 (0.028) −0.044 (0.005) 0.074 (0.033)
0.302 (0.016) −0.029 (0.01) −0.004 (0.005) 0.269 (0.015) −0.071 (0.011) 0.015 (0.003)
0.392 (0.013) 0.089 (0.009) −0.018 (0.005) 0.017 (0.012) −0.116 (0.012) 0.003 (0.002)
Leveraget −1
−0.052
−0.152
R-square F No. of observations No. of firms
(0.006) 0.048 78.38 33 873 2834
(0.005) 0.099 170.98 33 079 2834
RI Size Tangibility ROAt −1 Growth PRRI
3. Methodology and data 3.1. Variables and their measurement 3.1.1. Dependent variables Leverage ratio: The ratio of total debt to total assets. Long term debt maturity ratio (LTM): the ratio of long term debt to total debt is the long term debt maturity ratio (LTM). By following the Barclay and Smith approach8 (Barclay and Smith, 1995) debts with a maturity period of more than one year were considered as long term debts. Debt Specialisation Ratios (DSR): DSR is defined as the sum of the squared proportion of individual type of debts9 that constitute the total debt (Colla et al., 2011). 3.1.2. Independent and control variables The main independent variable in this study was the index of financial liberalisation adopted from Abiad et al. (2008). These researchers developed a database of financial reforms for many economies/countries to capture the gradual and stepwise developments in the following seven broad dimensions of regulation in the financial market: credit control and reserve requirements; external finance control; interest rate controls; bank entry barriers; bank privatisation; securities markets and bank supervision. Abiad et al. (2008) assigned a liberalisation score for each of the seven dimensions based on the extent of liberalisation, and then summed up the individual scores to arrive at the final score. Over the years, the index value changes based on the liberalisation measures adopted in any particular dimension. Detailed description of the index/database can be found in Abiad et al. (2008). As long as the liberalisation measure has sufficient time variance (and this variation is significant enough to prompt a response), it should be possible to conduct effective tests of the hypotheses laid out in the preceding section. 3.2. Model specification The base model used for estimation is presented in Eq. (1) Yjt = αj + β1 RI t + β2 Sizejt + β3 Tangibilityjt
+ β4 ROAjt −1 + β5 Growthjt + εj
(1)
7 Firms in developed financial markets have high degree of debt specialisation because firms choose that type of debt which suits its need the best (Rauh and Sufi, 2010). 8 Barclay and Smith (1995) consider debt with remaining maturity period of more than three years as the long term debt. But, they tested their hypotheses using debt with more than one year to maturity also. 9 We consider eight types of debt. They are bank debt, debt from financial institutions, government debt, loans from promoters, inter-corporate loans, deferred credit, trade credit and foreign currency borrowings.
0.142 292.46 33 873 2834
0.143 245.67 33 873 2834
Dependent variables: Leverage is the ratio of total debt to total assets. LTM is the ratio of long term debt to total debt, DSR is defined as the sum of the squared ratio of individual type of debt over total debt. Independent variables: RI is the financial reforms index adopted from Abiad et al. (2008), size is log (sales), tangibility is the ratio of net fixed assets to total assets, ROA is return on assets, growth is the annual growth rate in total assets. PRRI is an interaction dummy which is the product of RI and Priority dummy (Priority firms take the value zero and non-priority firms one). The parameters are estimated using the fixed effects estimation procedure. In parenthesis are the robust standard errors clustered at firm level.
where, Size: Following the methodology of Rajan and Zingales (1995) we measure size as the log of yearly sales. Growth: The annual growth rate in total assets is a growth proxy that has been used earlier by Fama and French (2001). Tangibility: The ratio of net fixed assets to total assets was used as a measure of tangibility. Profitability; To proxy for performance (Profitability) we use Return on Assets (ROA) which is defined as the ratio of profit before tax, interest and depreciation to total assets. Lagged values of ROA were used because size and ROA variables are endogenously related.10 The parameters were estimated using the fixed effects estimator.11 3.3. Data Firm level data was taken from the database of Centre for Monitoring Indian Economy (CMIE) which contains information of listed and unlisted firms from 1988 onwards. For the purpose of analysis, the present study used firm data which spans the period 1988 through 2005.12 The end point for data was dictated by the measure for liberalisation, viz. reforms index, which ended in 2005. Firms were chosen based on two criteria: (i) they should have existed before 1992, and (ii) they should have at least three years of leverage data for the period 1988–1992.13
10 Huselid et al. (1997) argue that firm size is one of the important determinants of firm performance. 11 Haushman test rejected the null at 1%. 12 Though our study period begins in 1988, the actual financial liberalisation took place, according to the International Monetary Fund in November, 1991. Thus, index value from 1988 to 1991 is very minimal. 13 This is required to classify firms into priority and non-priority firms.
36
N. Jadiyappa et al. / Economics Letters 149 (2016) 33–37 Table 2 Correlation matrix. Variables
Leverage
RI
Size
Leverage RI Size Tangibility ROAt −1 Growth
1.000 −0.125 −0.209 0.013 −0.256 −0.125
1.000 0.073 0.041 −0.150 −0.082
−0.125
Tangibility
ROAt −1
Growth
1.000 −0.084 −0.088
1.000 0.164
1.000
1.000 0.198 0.144
Leverage is the ratio of total debt to total assets. RI is the reforms index adopted from Abiad et al. (2008), size is log (sales), tangibility is the ratio of net fixed assets to total assets, ROA is return on assets, AGR is the annual growth rate in total assets.
4. Results and discussion 4.1. Hypothesis H1 and H1A In the first hypothesis, it was argued that financial reforms will have a significant impact on the leverage ratio of firms because of the reasons mentioned in the previous sections. The results are presented in Table 1. As hypothesised, the RI coefficient, which represents the reforms, is negative and significant, implying a negative impact of financial reforms on the debt level of firms. This result supports the findings of Bertrand et al. (2007) and Senay et al. (2007) who respectively observed the same results in the French and the developing world context. Bertrand et al. (2007) attribute this decrease to the abolition of government credit schemes in the post-reform period while Senay et al. (2007) ascribed it to the reforms which were aimed at minimising the impact of imperfections in the legal, capital and administrative spheres. In India as observed by Reddy (2002), the observed decline in debt ratios can be attributed to both the causes. These two reasons are complementary to each other and not mutually exclusive i.e., government loans would stop coming because of reforms and firms will switch to stock markets to fund their new growth.14 The next step involved analysis of the differential impact of financial liberalisation on priority and non-priority firms.15 Following the approach of Bertrand et al. (2007) firms were classified into priority and non-priority groups based on the median value of average long term leverage for the period 1988–1992. The results are presented in the third column of Table 1. The main variable of interest in this analysis was the product term (PRRI) which interacts the effects of financial reforms (RI) with the priority and nonpriority classification (PR). This interaction term gives a differential slope coefficient for non-priority firms.16 As is evident, the RI variable is negative and significant, indicating a negative impact on priority firms which is consistent with the study’s prediction. The interaction coefficient was found to be positive, but the adjusted slope coefficient was negative for non-priority firms as well. This is contrary to what had been predicted in H1A. However, it should be noted that the rate of decrease is significantly lower for nonpriority firms. 4.2. Hypothesis H2 Debt maturity and debt specialisation analysis As discussed earlier, financial liberalisation should exert a negative effect on LTM ratio. The results are presented in the fourth
14 As bank dominated financial system progress towards stock market dominated system, it is expected that firms start substituting debt by external equity to fund their investments. Consequently, debt level would decline. 15 Financial regulation in the pre-reforms period consisted of a complex web of rules and regulations. Any credible classification system should consider this complex web of regulations. In absence of information on such a complex system, we adopt an indirect method to classify firms into priority and non-priority firms as suggested by Bertrand et al. (2007). 16 Non priority firms take value one in PR dummy.
column of Table 1 and provide evidence to support the Hypothesis H2, because RI variable is negative and significant (−0.029). If there is truth in the argument that the negative impact is because of abolition of government support, then the observed negative impact in column four should be driven by the negative impact on the priority firms. This supposition was verified in an unreported analysis by including PRRI variable (interaction variable between priority dummy and RI) into Model III of Table 1. The results were found to be exactly as predicted with priority firms having a negative coefficient and non-priority firms having a positive adjusted coefficient. The findings in the present study contradict the result of Senay et al. (2007) who found no significant impact of financial reforms on the debt maturity ratio of firms in emerging countries. This might be due to differences in the starting points of financial reforms in various countries included in their sample, which is a very common problem in a cross-country study like theirs. 4.3. Hypothesis H3 Lastly, the results of the debt specialisation analysis are presented in the last column of Table 1. The debt specialisation ratio (DSR) that was used in our analysis is negatively related to debt diversification. That is to say, higher values of DSR indicate higher debt concentration; in other words, firms use a few dominant sources of debt. It is very clear that financial liberalisation caused an increase in the concentration of debt and a decrease in the diversification of debt sources as the RI co-efficient is positive and significant. On further analysis of this observation, it was found that the observed decrease in debt diversification is mainly due to substitution of informal sources of debt17 by the formal sources of debt like bank debt. Therefore, it can be stated that in the Indian context, financial reforms increased the access to formal sources of debt. 5. Conclusion This study shows how financial reforms could affect the behaviour of firms at the micro level. This result assumes great importance in the context of the introduction and implementation of financial reforms in many developing countries. The likely behavioural changes that these reforms would induce at the firm level are of great significance to both policy makers and investors alike. Since reform measures are converging across countries along the lines of developed world practices, the results from studies on individual countries could easily be generalisable across other countries too. It is very likely that the results of this study namely, a decline in debt/debt maturity and an increase in specialisation levels, would also be manifested in other countries which initiate reforms in their financial systems. Appendix A See Table 2.
17 Like inter-corporate loans, loans from promoters, friends, relatives,etc.
N. Jadiyappa et al. / Economics Letters 149 (2016) 33–37 Table 3 Regression results after clustering the standard errors at industry level.
References
Variables
Leverage Model I
Leverage Model II
LTM Model III
DSR Model IV
Intercept
1.171 (0.028) −0.266 (0.028) −0.093 (0.011) −0.048 (0.027) −0.622 (0.049) −0.064 (0.006)
1.183 (0.028) −0.113 (0.028) −0.094 (0.011) −0.090 (0.026) −0.627 (0.048) −0.064 (0.006) −0.282 (0.023)
0.139 (0.016) −0.072 (0.015) 0.018 (0.005) 0.465 (0.019) −0.031 (0.014) 0.030 (0.003)
0.357 (0.013) 0.088 (0.011) −0.024 (0.003) 0.016 (0.011) −0.102 (0.015) 0.002 (0.002)
0.011 (0.008) 0.151 102.720 33 873 2834
−0.086
RI Size Tangibility ROAt −1 Growth PRRI Leveraget −1 R-square F No. of observations No. of firms
0.123 180.790 33 873 2834
0.150 171.230 33 873 2834
(0.005) 0.054 81.370 33 079 2834
Dependent variables: Leverage is the ratio of total debt to total assets. LTM is the ratio of long term debt to total debt, DSR is defined as the sum of the squared ratio of individual type of debt over total debt. Independent variables: RI is the financial reforms index adopted from Abiad et al. (2008), size is log (sales), tangibility is the ratio of net fixed assets to total assets, ROA is return on assets, growth is the annual growth rate in total assets. PRRI is an interaction dummy which is the product of RI and Priority dummy (Priority firms take the value zero and non-priority firms one). The parameters are estimated using the fixed effects estimation procedure. In parenthesis are the robust standard errors clustered at industry level.
Appendix B See Table 3.
37
Abiad, Abdul, Tressel, Thierry, Detragiache, Enrica, 2008. A New Database of Financial Reforms. IMF Working Papers (08/266). Barclay, Michael J., Smith Jr., Clifford W., 1995. The maturity structure of corporate debt. J. Finance 50 (2), 609–631. Bertrand, Marianne, Schoar, Antoinette, Thesmar, David, 2007. Banking deregulation and industry structure: Evidence from the French banking reforms of 1985. J. Finance 62 (2), 597–628. Blundell-Wignall, A., Browne, F., 1991. Macroeconomic Consequences of Financial Liberalisation: A Summary Report, OECD Economics Department Working Papers, No. 98, OECD Publishing. http://dx.doi.org/10.1787/088742707378. Colla, Paolo, Ippolito, Filippo, Li, Kai, 2011. Debt Specialization. SSRN eLibrary. http://dx.doi.org/10.2139/ssrn.1520902. Demirgüç-Kunt, Asli, Maksimovic, Vojislav, 1999. Institutions, financial markets, and firm debt maturity. J. Financ. Econ. 54 (3), 295–336. http://dx.doi.org/10. 1016/s0304-405x(99)00039-2. Fama, Eugene F., French, Kenneth R., 2001. Disappearing dividends: Changing firm characteristics or lower propensity to pay? J. Appl. Corp. Financ. 14 (1), 67–79. http://dx.doi.org/10.1111/j.1745-6622.2001.tb00321.x. Huselid, M.A., Jackson, S.E., Schuler, R.S., 1997. Technical and strategic human resources management effectiveness as determinants of firm performance. Acad. Manag. J. 40 (1), 171–188. Jadiyappa, Nemiraja, Mitra, Amarnath, 2012. Financial Regulations, Structure of Financial Systems and Capital Structure, August 30, 2012. Available at SSRN: http://ssrn.com/abstract=2138660 or http://dx.doi.org/10.2139/ssrn.2138660. Khanna, Sushil, 1999. Financial reforms and industrial sector in India. Econ. Polit. Weekly 34 (45), 3231–3241. Rajan, Raghuram G., Zingales, Luigi, 1995. What do we know about capital structure? Some evidence from international data. J. Finance 50 (5), 1421–1460. no.. Rauh, Joshua D., Sufi, Amir, 2010. Capital structure and debt structure. Rev. Financ. Stud. 23 (12), 4242–4280. http://dx.doi.org/10.1093/rfs/hhq095. Reddy, Venugopal, 2002. Monetary and Financial Sector Reforms in India: A Practitioner’s Perspective. Senay, Agca, NicolÃ, Gianni De, Detragiache, Enrica, 2007. Financial Reforms, Financial Openness, and Corporate Borrowing: International Evidence. IMF Working Papers (07/186). Verma, Jayanth, 1998. Indian financial sector reforms: A corporate perspective. Vikalpa 23 (1), 27–38.