Interactive effect of product and international diversification on capital structure Daniele Monteforte*, Assistant Professor in Business Economics and Management University of Calabria, Dep. of Business Management, Ponte Bucci, cubo 3C, Campus of Arcavacata - 87036 Rende (CS) Italy

Raffaele Staglianò Ph.D. student at Toulouse University, Department of Finance (France), CRM, 2 rue du Doyen-Gabriel-Marty, 31042 Toulouse Cedex 9, France

Abstract This paper examines the link between product diversification, geographic diversification and capital structure for a panel of medium and large Italian firms. Results indicate that product and geographic diversification individually are positively related to capital structure, but the interactive variable between product and geographic diversification have a negative and significant coefficient. In contrast with previous empirical evidence, our findings support the hypothesis that complexity that comes from diversification reduce access to debt. When firms engage simultaneously in both product and geographic diversification strategies, agency costs of debt and asymmetric information problems may increase, thus reducing debt capacity.

Key words: product diversification , geographic diversification, capital structure, interaction effect. JEL classification: G30, G32

*Corresponding Author: Dr. Daniele Monteforte, Tel. +39.0984.492269, cell +39.3356296768 Fax +39 0984.492267, email: [email protected]

1

1 Introduction and aims Since the seminal work of Rumelt (1974), as pointed by Datta et al. (1992), the relationship between both product and geographic diversification on firm value has been one of the most extensively researched areas in the discipline of industrial organization, strategic management and finance. There is now considerable empirical literature that confirms the existence of a relationship between corporate diversification and firm value. However, there is no consensus on the direction of this relationship (Martin and Sayrak, 2003; Villalonga, 2003). In more recent years some authors have tried to investigate on the mechanisms that drive the creation or the destruction of value as a consequence of diversification strategies. Some of these studies have examined the relationship between geographic diversification and capital structure of firms (Michel and Shaked, 1986; Fatemi, 1988; Lee and Kwok, 1988; Burgman, 1996; Chen et al., 1997; Chuang, Chang and Li, 1999; Chkir and Cosset, 2001; Singh, Davidson, Suchard, 2003; Low and Chen, 2004); while some other studies have examined the relationship between product diversification and capital structure (Barton and Gordon, 1987, 1988; Kochhar and Hitt, 1998; Lowe, Naughton and Taylor, 1994; Singh, Davidson, Suchard, 2003; Low and Chen, 2004). Despite these efforts the empirical evidences on these subjects are still controversial. It is not still clear why the two dimensions of diversification, product and geographic, have different effects on corporate leverage (Singh, Davidson, Suchard, 2003) and why even for the same type of diversification strategy the results proposed from various authors are different. There is little published research dealing with the nature of the relationship between these strategic choice. Most of the empirical evidence on the interaction effect between product and international diversification on capital structure has been done using US data. For example, Chkir and Cosset (2001) and Singh et al (2003) find that leverage increases with both product and international diversification. In this study, we examine how diversification strategies influence financing decisions. To add to the existing literature on this topic, this article use a unique hand-collected data set on large Italian firms. 2

The 582 firm- year observations (see table 1) are dominated by firms that diversify in both product and geographic direction (78,7%). This is the most interesting aspect of our dataset and allows us to get a clear picture on diversification and capital structure relation. Our results show that when Italian firms pursue only product diversification or geographic diversification this generates an increase in leverage; while, in contrast with previous research, when they pursue simultaneously both geographic and product diversification strategies the interactive effect have a negative and significant coefficient. We argue that this could be due to the increase of complexity in internal management mechanisms inside the firm and to the increase of information asymmetries and agency costs that this complexity generates (e.g., Bodnar, Tang, and Weintrop, 1999): in a such context is much more difficult for debtholders to monitor the firm and would lead to a higher cost of debt and as a result decrease debt capacity. The paper is organized as follows. Section 2 describes background. Section 3 provides information on the sample, the methodology, and the variables used. Section 4 presents the results. The conclusions follows in Section 5.

2 Literature review 2.1 Effects of product diversification on capital structure Following the work of Levy and Sarnat (1970) and of Lewellen (1971) on M&A activity, a large number of studies (i.e. Kim and McConnell, 1977; Bergh, 1997) have investigated the potential financial benefits on firm value deriving from a diversifying strategy, especially in unrelated businesses (Amit and Livnat, 1988; Bergh, 1997). One possible explanation for these benefits comes from the reduction in the firm’s operating risk because of mutual financial support among the different business units, the well known "co-insurance effect". Combining business with cash flows that are not perfectly correlated can potentially reduce the 3

volatility of earnings and the costs of financial distress, thus increasing total stakeholders' value and reducing the cost of capital, with an overall impact on debt levels (i.e. Barton and Gordon, 1987, 1988; Lowe, Naughton and Taylor, 1994). The firm's increased debt capacity subsequently generates increased tax shields and correspondingly less taxes paid for the business conglomerate (Majd and Myers, 1987; Ferris et al., 2003). Moreover, while shareholders can achieve in a relatively simple way their desired level of risk through diversification, the same goal is more difficult to be achieved from lenders at a low cost (Amato and Remolona, 2003; Warga, 2004). For this and other reasons it is reasonable to conclude that co-insurance effect could thus let lenders to accept lower returns on their loans, further improving debt capacity. At the same time, if it's true that unrelated diversification may allow more total external financing to be raised than could be raised by the individual businesses operating as standalones, thus mitigating the under-investment problem, it's at the same time true that this could exacerbate the over-investment problem with negative consequences on value (Scharfstein, 1998; Rajan et al., 2000). Another theoretical argument in support of financial benefits that could justify a diversifying strategy is the possibility to create an internal capital market (Williamson, 1975, 1986; Stein, 1994, 1997; Matsusaka et al, 2002) that can be useful in reducing information asymmetries between managers and investors and in providing for a better allocation of funds across projects within the firm. According to Williamson (1975) internal capital markets in diversified firms could allocate capital more efficiently than external market, in particular in case of capital rationing (Henderson, 1979). Khanna and Palepu (1997; 2000) argue that diversification should be more valuable in emerging markets because of information asymmetries, market imperfection and a lack of efficient governmental and financial institutions that make it more difficult for focused firms to find financing sources and to survive. Firms can take advantage of these imperfections by diversifying at the firm level or through membership in industrial groups that are common in many emerging and developed capital markets. There is a large 4

literature that reaches opposite conclusions both for developed and emerging economies. For example, Lamont (1997), Houston, James, and Marcus (1997), Shin and Stulz (1998), Scharfstein (1998), Scharfestein and Stein (2000) and Rajan, Servaes, and Zingales (2000) have demonstrated in the US that internal capital markets are inefficient in comparison to external capital market due to incentive and information problems that lead to cross-subsidization. There could be a potentially higher risk of value loss from diversification in emerging markets, since are often characterized by weak institutions, poor investor protection and poorly functioning capital, labor and product markets (Khanna and Yafeh, 2007). According to La Porta, Lopez de Silanes, Shleifer, and Vishny (1999) and Claessens, Djankov, and Lang (2000) in emerging markets there is a large divergence between cash flow rights and control rights and a widespread use of pyramids and business groups. This could lead majority shareholder to tunnel (Johnson et al., 2000), or exploit, resources away from minority investors. Even these conclusions, though, have not a wide consensus and there is a need for further investigations. For example, Chuhan et al. (1998), using the Worldscope database over 1991–1996 period for 9 Asian countries, provide evidence of a positive relation between per capita GNP and the valuation effects of both vertical and related diversification. Masulis et al (2009) empirically provides demonstrates the potential benefits of pyramids in reducing cost of capital and increasing firm value. And the literature on business groups is equally ambiguous since, as suggested by Khanna and Yafeh (2007), in some countries business groups are paragons, whereas in others they function like parasites. A third justification for the relationship between diversification and debt usage is of agency nature, due to the separation of ownership and control in many corporations. According to Jensen (1986) firm with low dividend payouts and low levels of debt obligations have more "discretionary cash flows" that could lead managers to adopt investments into low return project through diversifying strategies. Diversification may benefit managers because of the power and prestige associated with managing a larger firm (Jensen, 1986; Stulz, 1990), because managerial compensation is related to firm size 5

(Jensen and Murphy, 1990), because diversification reduces the risk of managers’ undiversified personal portfolio (Amihud and Lev, 1981), or because with a higher level of diversification is more complex to replace the actual management (Shleifer and Vishny, 1989). According to these and other considerations, shareholders should promote the use of debt so as to reduce agency costs (monitoring and discipline effect) but in this case the sign of the relationship between diversification, in particular unrelated diversification, and debt is negative. Also this hypothesis, though, has not received unanimous approval. In fact, some authors argue that diversification decreases the agency costs of debt because lower cash flow volatility decreases managerial discretion when the firm issue debt (Stulz, 1990), because of capital markets creation associated with diversification, and because of a better allocation of debt raised inside a diversified firm (Gertner, Scharfstein, and Stein, 1994). Finally, according to the transaction cost hypothesis (Alonso, 2003) the adoption of an unrelated diversification strategy indicates the presence of an excess of non-specific assets in the firm (Chatterjee and Wernerfelt, 1991). Since debt is the preferred source of financing when firm assets are not specific, according to this hypothesis firms following related diversification strategies are likely to be mainly equity-financed, while firms that follow unrelated diversification may prefer debt financing (Alonso, 2003).

2.2 Effects of geographic diversification on capital structure Early literature on the link between geographic and capital structure was unanimous to suggest that multinational firms (MNCs) are able to support higher levels of debt (Hughes, 1975; Rugman, 1976). Since then, the empirical results produced in literature have been contradictory and let open space to future research: many studies based on US have, in fact, demonstrated that US based MNCs have lower debt to equity ratio than their domestic counterparts (Lee, 1986; Fatemi, 1988; Lee and Kwok, 1988; Burgman, 1996; Chen et al., 1997) while other based on MNCs operating outside US, and especially 6

those located in emerging countries, demonstrated that there is a positive relationship between international activity and capital structure (Singh and Nejadmalayeri, 2004; Mitto, Zhang, 2005). As for product diversification, one of the main factor that influence capital structure decisions for MNCs is the business risk. In the case of internationally diversified firms, the imperfect cash flow correlations among different markets around the world may have a positive impact on risk (i.e., the cost of financial distress, expected bankruptcy cost, earnings volatility, etc) and, consequently, on leverage (Shapiro, 1996; Eiteman et al., 1998). Even though empirical evidence suggests that US MNCs have lower debt ratio than DCs (Lee, 1986; Fatemi, 1988; Lee and Kwok, 1988; Burgman, 1996), some authors (Chen et al, 1997; Chkir and Cosset, 2001) have demonstrated that their leverage increase with the degree of foreign involvement. The factor that has been addressed in literature as responsible for increase or decrease of financial risk is mainly the impact of foreign exchange risk and of social and political risks (Aliber, 1984): while some argue that foreign exchange and political risk has a negative impact on overall risk and on debt levels (Lee and Kwok, Burgman, 1996) others sustain the exact contrary (Chkir and Cosset, 2001). Recently, Kwok and Reeb (2000) propose a solution for this argument suggesting that capital structure of MNCs can differ between developed countries based and emerging countries based firms. They also provide empirical evidence that international diversification is negatively related to leverage for US based firms and positively related to leverage for emerging market-based firms. Finally when financial markets are not well integrated, Doukas and Pantzalis (2003) argue that MNCs could raise more capital through foreign debt financing and at more favourable terms than domestic firms and Thomadakis and Usmen (1991) show that foreign risky debt can increase shareholders value. Geographic diversification can also contribute to the creation of internal capital markets which leads, especially where external capital markets are inefficient, to cheaper sources of financing, to lower information asymmetries and higher debt levels, compared to external markets and to domestics firms 7

(Caves, 1971; Maksimovich and Philips, 2000; Doukas and Pantzalis, 2003). Empirical evidence on the subject, though, is quite mixed. Finally, however, geographic diversification is also associated in many studies to higher agency costs and, theoretically, to lower debt levels. For example, in the diversification literature there are numerous studies on the matter of the exacerbation of agency costs due to capital misallocations and inefficient cross-subsidization of poorly performing units inside diversified firms (Rajan, Servaes and Zingales, 2000; Scharfstein and Stein, 2000). Furthermore, according to agency literature managers are willing to create larger internal capital markets through diversification so as to avoid market discipline that comes with external financing (Datta, Mello and Datta, 2009) and to engage in activities that maximize their private benefits (Berger and Ofek, 1995) such as such as prestige, power, and compensations (Jensen, 1986; Stultz, 1990, Jensen and Murphy, 1990; Datta, Mello and Datta, 2009). Another agency related matter is the increase of complexity that comes with diversification. According to Bodnar, Tang and Weintrop (1999) globally diversified firms are inherently more complex than purely domestic firms and consequently the managers of such firms are more difficult for shareholders to monitor. According to this argument, diversification should be associated to lower debt levels and vice-versa. For the same reasons, difficulties in gathering and processing information for geographically dispersed MNC's make monitoring more costly also for bondholders and this could lead to higher interest payments on loans and, consequently to lower debt ratios than pure domestic firms (Doukas and Pantzalis, 2003).

2.3 Combined Effect of product and geographic diversification on capital structure A large number of studies have investigated the potential advantages ((Buhner, 1987, Hitt et al., 1997; Hill, Hitt and Hoskisson, 1992; Teece, Pisano and Shuen, 1997; Kogut and Zander, 1992) and costs (Hoskisson and Hitt, 1988; Hitt at al., 1997; Ruigrok and Wagner, 2003) of implementing 8

simultaneously both types of diversification and the overall effect that a combination of these two strategies have on firm performance (Chang, Wang, 2007). According to resource based view and organizational learning theory, the experience gained with product diversification can build managerial capabilities that help to manage more effectively some of the complex challenges posed by international diversification (Hitt et al., 1997); moreover, international diversification could offer prospective market opportunities (Buhner, 1987), for example giving to industrially diversified firms that decide to diversify internationally more opportunities to gain from the economies of scope and scale. Industrially diversified firms could also gain through global networks an improved efficiency in resources allocation (Porter, 1985) and greater competitive power (Kogut, 1984). On the contrary, according to the agency view (Jensen and Meckling,1976), the more complex the corporation, the more difficult is it for outsiders to monitor management’s decisions: when firms are diversified in product and geographic markets, agency costs of debt and asymmetric information problems can be greatly magnified to the extent that it becomes too costly for debtholders to have an adequate understanding of managerial decisions. Moreover, Hoskisson and Hitt (1988) argue that to manage large product and internationally diversified firms require newer structural forms and more complex (and innovative) control systems and Hitt et al. (1997) argue that as the number of foreign transactions increases, difficulties in coordination could increase, especially for firm engaged also in product diversification, increasing the level of complexity among the firm. Nevertheless, the empirical evidence on the interaction effect between product and international diversification has been mixed: while some authors (e.g., Hitt et al, 1997) argue that product diversification positively enhances the performance of internationally diversified firms, others (Geringer, Tallman, and Olsen, 2000; Tallman and Li, 1996) found no evidence for an interactive effect. At the same time, while there is a large number of studies that have investigated the separate effects of 9

industrial and geographic diversification on leverage, fewer are the studies on the combined effects of both types of diversification on capital structures across countries (Low, Chen, 2004). Published research dealing with the impact on leverage of the interrelationship between product and geographic diversification is very limited. Chkir and Cosset (2001) have examined the relationship between the debt level of MNCs and their diversification strategy, integrating into the analysis both the international market and the product dimension of diversification by means of a switching regression model that allows the effect of the determinants of the capital structure of MNCs to vary with the strategy of diversification. Their results suggest that leverage increases with two types of diversification and that the combination of both these types of diversification leads to lower levels of default risk for MNC's. Moreover, according to Chkir and Cosset (2001) the combination of both types of diversification enables MNCs to achieve higher levels of profitability than those of MNCs pursuing a single diversification strategy. Singh et al. (2003) considering a sample of U.S. listed firms, have found that firms following a strategy of dual diversification, product as well as international, are able to support higher leverage since the two diversification types seem to complement each other in creating debt capacity, while those who pursue individually only one type of diversification have a negative impact on leverage from diversification. We address several questions. First, we re-examine the impact of product and geographic diversification and of interaction between these diversification strategy on debt capacity. Second, with the objective to verify the efficiency of diversification strategy, we examine these relations sorting the sample by firm performance.

3 Methodology 10

3.1 Data The analysis is based a database provided by Ricerche and Studi of Mediobanca on the largest financial and non-financial Italian companies, including holding companies with consolidated balance sheets. Financial companies were excluded. We consider five years of data (2003–2007). The sample varies over time, on average it includes 115 large companies (43% of the companies are listed in the Italian stock exchange); on the whole we have 582 observations. Because of missing accounting data in some years for some firms the number of observations in some regressions is reduced. The number of observation is reduced also for empirical specification that included lagged variables. The dataset includes balance-sheet data, as well as information on product and international diversification. Since only a part of the companies is listed in the Italian stock market, we cannot use information on market valuation. Sample includes a very limited number of focused firms that we exclude from the analysis.

3.2 Variables Dependent Variables. Firm leverage, or the level of debt financing (TOTDEBT_TA) was operationalized as the ratio of total debt to total assets of the firm (Singh et al., 2003). Explanatory variables. Being able to compare our results with those in the literature, we used the entropy index to estimate product diversity (PROD_DIV). This measure has been widely used in measuring product diversification in the literature (e.g., Hitt et al., 1997; Jacquemin & Berry, 1979; Palepu, 1985). PROD _ DIV = ∑ i Pi ln (1 / Pi )

where P i is the proportion of sales attributed to product segment “i”. To measure international diversify, we use sales of the ratio of foreign sales to total sales (INT_DIV).

11

(e.g., Gaur and Kumar, 2009; Singh et al.,2003). Some previous studies used the entropy approach to measure international diversity (e.g., Hitt et al., 1997). We can not use this measure because in the database we don’t find the same methods to group foreign markets. As previously observed, many of the product-diversified firms are also geographically diversified. To determine if there is interaction between the two types of diversification in their impact on capital structure we consider the interaction variable (INT_PI). Control variables. There is mixed support for firm-characteristics determinants of capital structure (e.g., Balakrishnan and Fox, 1993; Lim et al., 2009). The control variables are firm size (natural logarithm of net sales),SIZE,

tangibility (ratio of plant, property and equipment to net sales),

TANGIBILITY, performance (EBIT/ total assets), PERFORMANCE, age (natural logarithm of the number of years since the date of its foundation), AGE, and finally a dummy variables, D_LISTED, that takes value 1 if firm is listed. In the database we have information only for one year on ownership structure. The correlation between ownership concentration of largest shareholders and D_LISTED is very large (in mean =- 0.65). We can consider the variable D_LISTED as a proxy of increasing level of number of shareholder that hold a security. The effect of this variable is similar when a firm increases the dispersion of its ownership. Finally, we control for year dummies and to control for industryspecific effects, we follow Pavitt’s taxonomy (1984).

3.3 Empirical specification Multivariate regression models are used to relates debt to product and international diversification, as well as a number of control variables which are discussed previously. the basic model is given as:

TOTDEBT_TA = f [ INT _ DIV , PROD _ DIV , INT _ PI , SIZE , TANGIBILITY , PERFORMANCE , AGE , D _ LISTED, YEAR DUMMIES , INDUSTRY DUMMIES

]

(1)

12

Empirically, we first take advantage of a panel data set and use a fixed-effect estimator in equation (1) to control for unobservable firm characteristics, year and industry effects. This approach reduce or avoid bias with respect to omitted variables. The econometric technique used in the model included the computation of heteroskedasticity-consistent standard errors. Second, we accounts for the potentially dynamic nature of a firm’s debt considering the lagged dependent variable as an additional regressor. From an interpretative point of view, Fischer et al. (1989) and Flannery and Rangan (2006), suggests that firms follow leverage targets. Firms actively rebalance capital structure in order to close the gap between the current and the targeted leverage.

TOTDEBT_TA = f [TOTDEBT_TA (t - 1), INT _ DIV , PROD _ DIV , INT _ PI , SIZE , (2) TANGIBILITY , PERFORMANCE , AGE , D _ LISTED, YEAR DUMMIES , INDUSTRY DUMMIES ]

We use two econometric methods for dynamic relation. The two econometric methods are the fixed effect estimation and a dynamic panel model with bias-corrected fixed effect estimator. We apply the bias-corrected fixed effect estimator because considering the lagged dependent variable implies obvious problem of endogeneity. A natural solution for first-order dynamic panel data models is to use GMM (General Method of Moments; see Anderson and Hsiao, 1982; Arellano, 1989; Arellano and Bond, 1991; Arellano and Bover, 1995; Blundell and Bond, 1998). Unfortunately, this method is only efficient asymptotically and is not suitable for small samples (such as the ones used here). We run a corrected version of the fixed effects model in STATA, using the “xtlsdvc” command developed by Bruno (2005) for unbalanced panels. This method has been proposed as a suitable panel data technique in the case of small unbalanced samples where GMM cannot be applied efficiently (Kiviet, 1995; Judson and Owen,1999; Bun and Carree, 2005).We used Anderson and Bond as the first step estimator.

13

An alternative procedure to assess the efficiency of diversification strategy (both product and international) is to evaluate the impact of interaction variable (INT_PI) on debt capacity sorting the sample by firm performance. Thus we would expect that the negative impact of INT_PI on leverage tends to be statistically significant only for low-performing firms. If our general results are descriptive of profit-reducing behaviour, the firms which follow the predictions more closely should have worse performance.

4 Results 4.1 Descriptive Statistics Table 2 provides mean and median value for continuous variables for our full sample of diversified firms. We report t-statistic for differences in the means. We find that debt to asset ratios are highest for firms that diversify in both directions. This average is significantly only with respect geographic diversified firms (p-value=0.0188). Multinational multi-activity firms dominate, in term of mean, geographic diversified firms considering the variables SIZE, TANGIBILITY, PERFORMANCE and AGE. Product diversified firms are larger than multinational multi-activity in terms of SIZE and TANGIBILITY. In contrast, product diversified firms are, on average, significantly lower than firms that diversify in both direction with respect the variable PERFORMANCE. To sum-up, the evidence suggests that the Italian firms with the highest performance diversify in both directions. Debt ratio is less conclusive, with no significant difference between product diversified firms and multinational multi-activity firms.

---Here table 2 ---

Table 3 shows the correlation matrix between all variables analysed. 14

---Here table 3 ---

Results exhibit the marginality of all the correlation parameters, which does not bias the statistic significance of results produced by using this sample. Tests of Variance Inflactor Factor (VIF), not reported, is close to one in all the cases confirming that the absence of multicollinearity can be accepted.

4.2 Regression Results Table 5 displays the results of leverage regressions based on model (1). Regression 1 and 2 consider full sample. In regressions 3 we consider only multinational multi-activity firms (INT_PI>0). First, in regression 1 we include only diversification variables (regression 1) while regression 2 and 3 incorporate also control variables.

---Here table 4 ---

In all regressions the variables INT_DIV and PROD_DIV are positively related to the leverage, and the coefficients are all significant. Interestingly, the coefficient on our interaction variable INT_PI is statistically negative in all regressions. The results suggest that firms that diversify in both direction have a limited debt capacity while individually, they may be positively related to firm leverage. The empirical indicate that when firms engage in both diversification strategies, agency costs of debt may increase reducing debt capacity. The complexity created from both type of diversifications, may increase information asymmetries and agency costs that would lead to a lower debt capacity. We also find that SIZE and TANGIBILITY are negatively related to leverage while the coefficient of 15

PERFORMANCE is in all regressions statistically negative. Table 5 shows the results considering the lagged dependent variable, TOTDEBT_TA (t-1), to accounts for a dynamic component in leverage. Regressions contains the same set of explanatory and control variables considered in table 4. ---Here table 5 ---

Consistent with previous findings, in general, the evidence suggest that the interaction variable INT_PI is significantly and negatively related to the debt ratios in all regressions. We also find that the coefficients of variable INT_DIV and PROD_DIV are also in these cases significantly positive. The sign of the coefficient for lagged leverage is significantly positive in all regressions. More leveraged firms have more debt capacity. Column (3) produces results for coefficients that are corrected for dynamic panel bias using bias corrected fixed effects model. In comparison to uncorrected coefficient values, the results do not change substantially. Table 6 displays the results for our sub-group analysis. The sample was divided into two depending on whether they were above or below the mean performance measure. The sub-groups with values above the mean of PERFORMANCE characterise high- performing firms, while sub-groups with values lower than mean of PERFORMANCE characterise low-performing firms.

---Here table 6---

The results show that the low-performing firms conform to our model better. In particular, the coefficients of the interaction variable, INT_PI, is significant for the low-performing firms and for the high-performing are not.

16

5 Conclusions This study contributed to international finance and strategy researchs by demonstrating the systematic influence of product and international diversification on debt capacity. Using a unique hand-collected data set, the results presented in this paper suggest interactive variable between product and geographic diversification have a negative and significant coefficient. This article, suggest that the Italian model of finance may produce different results from its Anglo-Saxon counterpart. For example, our results are in contrast with Singh et al. (2003) that for US listed firms report positive coefficients for interactive terms. Further, when we consider only multinational multi-activity firms we find that these firms have low debt capacity. These results are robust to different empirical specifications. Using static or dynamic firm-level panel data model does not change the direction of results. Finally, we also found that lowperforming firms confirm the general results better than high-performing firms. These results support the hypotheses that when firms diversify in product and geographic markets may increase coordination problems and information processing (e.g., Hitt et al.,1997). The complex structure of these firms may exacerbate asymmetric information problems and agency costs of debt implying a reduction of debt ratio.

17

References Amato, J.and E. Remolona (2003): “The Credit Spread Puzzle”, BIS Quarterly Review, dicembre, pagg. 53-66. Amihud, Y. and B. Lev, (1981), "Risk Reduction As A Managerial Motive For Conglomerate Mergers," Bell Journal of Economics, 12, 605-617. Amit, R. and J. Livnat (1988): “Diversification and the risk-return trade-off”. Academy of Management Journal, Vol. 31, pp. 154–166. Anderson, T. W. and C. Hsiao. (1982). Formulation and estimation of dynamic models using panel data. Journal of Econometrics 18: 570{606. Arellano, M. (1989) A Note on the Anderson-Hsiao Estimator for Panel Data, Economics Letters, vol.31, pp. 337-41. Arellano, M. and S. Bond. 1991. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies 58: 277{297. Arellano, M. and Bover, O. (1995) Another Look at the Instrumental Variables Estimation of Error components Models, Journal of Econometrics, vol. 68, pp.29-51. Balakrishnan S, Fox I. 1993. Asset specificity, firm heterogeneity and capital structure. Strategic Management Journal 14(1): 3–16. Barton, S., and P. Gordon, 1987, Corporate strategy: useful perspective for the study of capital structure? Academy of Management Review 12, 67–75. Barton, S., and P. Gordon, 1988, Corporate strategy and capital structure, Strategic Management Journal 9, 623–632. Berger, Philip G., and Eli Ofek, 1995, Diversification’s effect on firm value, Journal of Financial Economics 37, 39-65. Bergh, D. (1997) ‘Predicting divestiture of unrelated acquisitions: An integrative model of ex-ante 18

conditions’, Strategic Management Journal, 18:715-731. Blundell, R. and S. Bond. 1998. Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics 87: 115{143. Bodnar, Gordon M., Charles Tang, and Joseph Weintrop, 1999, Both sides of corporate diversification: The value impacts of global and industrial diversification, Working paper, Johns Hopkins University. Bruno, G. S. F. (2005). Approximating the Bias of the LSDV Estimator for Dynamic Unbalanced Panel Data Models. Economic Letters 87, 361–336. Buhner, R. (1987). Assessing International Diversification of West German Corporations. Strategic Management Journal, 8(1), 25-37. Bun, M. J. G. and M. A. Carree (2005). Bias-Corrected Estimation in Dynamic Panel Data Models. Journal of Business and Statistics 23(2), 200–210. Burgman, Todd A., 1996. An empirical examination of multinational corporate capital structure. Journal of International Business Studies 27(3), 553–570. Chang, S. & Wang, C. (2007). “The effect of product diversification strategies on the relationship between international diversification and firm performance”. Journal of World Business, 42: 61–79 Chava, S., P. Kumar and A. Warga, 2004, Agency Costs and the Pricing of Bond Covenants. Working paper, University of Houston. Chen, Charles J.P., Cheng, Anges C.S., He, Jia, Kim, Jawon, 1997. An investigation of the relationship between international activities and capital structure. Journal of International Business Studies 28(3), 563–577. Chkir, I. E. and J. Cosset, “Diversification Strategy and Capital Structure of Multinational Corporations.” Journal of Multinational Financial Management (Feb.), 17–37 (2001). Chuang, C., T. Chang and P. P. Li, “The Internationalization and Capital Structure of Taiwanese 19

Multinationals.” American Asian Review 17(Winter), 51–75 (1999). Chuhan, Punam, Stijn Claessens and Nlandu Maimingi (1998): “Equity and bond flows to Latin America and Asia: the role of global and country factors”, Journal of Development Economics, vol 55, pp 439–63. Datta, D. K., N. Rajagopalan and A. Rasheed (1991). 'Diversification and performance: Critical review and future directions', Journal of Management Studies, 28, pp. 529-558. Datta, S., D'Mello, R. and Iskandar-Datta, M. 2009, "Executive compensation and internal capital market efficiency", Journal of Financial Intermediation, vol. 18, pp. 242-258. Doukas, J., Pantzalis, C., 2003. Geographic diversification and agency costs of debt of multinational firms. Journal of Corporate Finance 9, 59–92. Fatemi, Ali M., 1988. The effect of international diversification on corporate financing policy. Journal of Business Research 16(1), 17–30. Ferris, S., Jagannathan, M. and Pritchard, A. C. (2003), “Too busy to mind the business? Monitoring by directors with multiple board appointments”, The Journal of Finance, Vol. 58, pp. 1087-1111. Fischer, E., Heinkel, R., Zechner, J., 1989. Dynamic capital structure choice: theory and tests. Journal of Finance 44, 19–40. Flannery, M. J. and Rangan, K. P. (2006) ‘Partial adjustment toward target capital structures’, Journal of Financial Economics, Elsevier, 79(3), 469--506. Gaur Ajai S. and Vikas Kumar International Diversification, Business Group Affiliation and Firm Performance: Empirical Evidence from India, British Journal of Management, Vol. 20, 172–186 (2009) Geringer, J. M., Tallman, S., & Olsen, D. M. (2000). “Product and international diversification among Japanese multinational firms”. Strategic Management Journal, 21: 51–80. Hill, C. W. L., Hitt, M. A., & Hoskisson, R. E. 1992. Co-operative versus competitive structures in 20

related and unrelated diversified firms. Organization Science, 3: 501-21. Hitt, Michael A., Hoskisson, Robert E., Kim, Hicheon, 1997. International diversification: effects on innovation and firm performance in product-diversified firms. Academy of Management Journal, 40(4), 767–798. Hoskisson, R.E. & Hitt, M.A. 1988. Strategic control systems and relative R&D investment in large multiproduct firms. Strategic Management Journal, 9: 605-621. Jacquemin, A. P. and Berry, C. H. (1979) Entropy Measure of diversification and corporate growth, The Journal of Industrial Economics, 4: 359–69. Jensen M., (1986), “Agency Costs of Free Cash Flow, Corporate Finance, and Takeover”, American Economic Review, vol. 76. Jensen, M., Murphy, K. (1990), Performance pay and top-management incentives, Journal of Political Economy, 98. Johnson, Simon, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer, 2000, Tunneling, American Economic Review 90, 22-27. Judson, R. A. and Owen, A. L. (1999) Estimating Dynamic Panel Data Models: A Guide for Macroeconomists, Economics Letters, vol.65, pp. 9-15. Khanna, T. and Palepu, K. (2000a). Is Group Affiliation Profitable in Emerging Markets? An Analysis of Diversified Indian Business Groups. Journal of Finance, pages 867–891. Khanna, T. and Yafeh, Y. (2007). Business Groups in Emerging Markets: Paragons or Parasites? Journal of Economic Literature, 45(2):331–372. Kim, E. Han, and John J. McConnell, 1977, Corporate mergers and the co–insurance of corporate debt, Journal of Finance 32, 349–365. Kiviet, J. F. (1995) On Bias, Inconsistency, and Efficiency of Various Estimators in Dynamic Panel Data Models, Journal of Econometrics, vol.68, pp. 53-78. 21

Kochhar R, Hitt MA. 1998. Linking corporate strategy to capital structure: diversification strategy, type and source of financing. Strategic Management Journal 19(6): 601–610. Kwok, C.C. and Reeb, D.M. (1998) “Internationalization and firm risk: an upstream and downstream hypothesis”, Journal of International Business Studies, 31(4):611-629. Lamont, O., 1997. Cash flow and investment: evidence from internal capital markets. «Journal of Finance» 52, 83–109. Lee, Kwang Chul, Kwok, Chuck C.Y., 1988. Multinational corporation vs domestic corporation: international environment factors and determinants of capital structure. Journal of International Business Studies 19(4), 41–58. Levy, H. And M. Sarnat, “Diversification, Portfolio Analysis and the Uneasy Case for Conglomerate Mergers”, Journal of Finance, 25 (4), 1970, pp. 795-802. Lewellen, W., (1971), "A Pure Financial Rationale For The Conglomerate Merger," Journal of Finance, 26, 521-537. Lim Elizabeth Ngah-Kiing * Shobha S. Das,2 And Amit Das2, Diversification Strategy, Capital Structure, And The Asian Financial Crisis (1997–1998): Evidence From Singapore Firms, Strategic. Management Journal., 30: 577–594 (2009) Low, P., and K. Chen, 2004, Diversification and capital structure: some international evidence, Review of Quantitative Finance and Accounting 23, 55–71. Lowe, Julian, Naughton, Tony and Taylor, Peter. The impact of corporate strategy on the capital structure of Australian Companies. Managerial and Decision Economics, Vol. 15, No. 3. (May – Jun., 1994), pp. 245-257. Maksimovic, V., and G. M. Phillips, 2000, “Optimal Firm Size and the Growth of Conglomerate and Single-Industry Firms,” Journal of Finance, 55, 721—767. 22

Martin, J. and Sayrak, A. 2003, ‘Corporate diversification and shareholder value: a survey of recent literature’, Journal of Corporate Finance, 9, 37 – 57. Matsusaka, John, and Vikram Nanda, 2002, Internal capital markets and corporate focusing, forthcoming, Journal of Financial Intermediation 11, 176-211. Majd, S. and Stewart Myers (1987) “Tax Asymmetries and Tax Reform.” in: Martin Feldstein (ed.) Effects of taxation on capital accumulation, University of Chicago Press, Chicago IL. Michel, Allen and Shaked, Isreal (1986) “Multinational corporations verses domestic corporations: Financial performance and characteristics” Journal of International Business Studies, p. 89-100. Pavitt, K. (1984). "Sectoral patterns of technical change: towards a taxonomy and a theory". Research Policy 13: 343–373. Rajan, R., H. Servaes, & L. Zingales, (2000), “The cost of diversity: The diversification discount and inefficient investment”. Journal of Finance, 55, 35-80. Rajan, Raghuram G., Henri Servaes, and Luigi Zingales, 2000, The cost of diversity: The diversification discount and inefficient investment, Journal of Finance 55, 35-80. Ruigrok, W., & Wagner, H. 2003. Internationalization and performance: An organizational learning perspective. Management International Review, 43: 63-73. Rumelt R.P. (1974), Strategy, Structure and Economic Performance, Cambrige, MA, Harvard University Press. Scharfstein, David S., 1998, The dark side of internal capital markets II: Evidence form diversified conglomerates, Working paper, Massachusetts Institute of Technology. Shleifer, A. &

R. Vishny, (1989), “Managerial entrenchment: The case of manager-specific

investments”, Journal of Financial Economics, 25, pp. 123-139. Singh, M., Davidson,W.N., Suchard, J., 2003. Corporate diversification strategies and capital structure. Quarterly Review of Economics and Finance 43, 147–167. 23

Singh, M. and Nejadmalayeri, A. (2004) “Internationalization, capital structure, and cost of capital: evidence from French corporations”, Journal of Multinational Financial Management, 14:153-169. Stein, Jeremy C., 1997, Internal capital markets and the competition for corporate resources, Journal of Finance 52, 111-133. Stulz, R.M., (1990), Managerial discretion and optimal financing policies, Journal of Financial Economics 26, 3-27. Tallman, S., & Li, J. T. (1996). “Effects of international diversity and product diversity on the performance of multinational firms”. Academy of Management Journal, 39(1): 179–196. Thomadakis, Stavros and Nilufer Usmen, 1991, Foreign project financing in segmented capital markets: Equity versus debt, Financial Management 20 (4), 42-53. Villalonga, Belén (2003), “Research Roundtable Discussion: The diversification discount,” Financial Economics Network FEN-Educator Series, Social Science Research Network (SSRN), Williamson, O. Markets and Hierarchies: Analysis and Antitrust Implications. New York, NY: Free Press,1975. Williamson, O., 1986, Economic organization: Firms, markets and policy control. New University Press, New York.

24

Table – 1- Firms’ diversification status in years 2003-2007

Diversification Status

No.OBS (No.firms)

Only product diversified

71 (15)

No.OBS in % (No.firms in %) 12.20% (12.20%)

Only Geographic diversified

53 (12)

9.10% (9.75%)

Diversified in both directions

458 (96)

(78.70%) (78.05%)

Total

582 (123)

100 (100)

25

Table – 2- Mean comparison across firms and firms’ diversification status Year

Only product diversified (1)

Diversified in both directions (3) .6513459 (.6710702) 20.94668 (20.79788)

(1) vs (2) (p-value)

(1) vs (3) (p-value)

(2) vs (3) (p-value)

.6485914 (.6331925) 21.04965 (21.08304)

Only Geographic diversified (2) .5888643 (.6429327) 20.34496 (20.33466)

TOTDEBT_TA

0.059727 (0.0490) 0.704699 (0.0001)

-.0027545 (0.9011) 0.1029696 (0.4354)

-.0624815 (0.0188) -.6017294 (0.0000)

TANGIBILITY

.9144504 (.7822633)

.1927412 (.1797386)

.3407894 (.2056555)

0.7217092 (0.0000)

0.5736609 (0.0000)

-.1480483 (0.0330)

PERFORMANCE

.0548082 (.0536134)

.0397634 (.0366643)

.069482 (.0623079)

0.0150448 (0.1143)

-.0146737 (0.0697)

-.0297186 (0.0021)

AGE

2.95037 (2.890372)

3.140235 (3.135494)

3.183317 -0.189865 (3.433987) (0.2892)

-0.2329461 (0.0917)

-.0430813 (0.7755)

SIZE

The first row of each variable represents the mean, the second row (in bracket) represents the median. Differences significance of the means are tested using the two sample t-test.

26

Table 3– Variable description and correlation matrix

1 TOTDEBT_TA 2 INT_DIV 3 PROD_DIV 4 SIZE 5 TANGIBILITY 6 PERFORMANCE 7 D_LISTED 8 AGE

1 1 0.0669 0.0224 0.1764* -0.0728 -0.2448* 0.0026 -0.0794

2

3

4

5

1 -0.1103* -0.0803 -0.3586* -0.0457 -0.0582 -0.0129

1 0.3123* 0.0733 0.0528 0.2289* 0.0872

1 -0.0334 0.191* 0.2555* -0.0443

1 -0.1439* 1 0.2133* -0.0689 1 -0.178* 0.0743 -0.055 1

6

7

8

(*) indicates a level of significance lower than 5%.

27

Table 4 – Leverage and diversification: Fixed effect static models

VARIABLES

INT_DIV PROD_DIV INT_PI

FE

FE

FE2 (only obs with INT_PI>0) (1) (2) (3) TOTDEBT_TA TOTDEBT_TA TOTDEBT_TA

0.3783*** (0.087) 0.1264*** (0.048) -0.2906*** (0.094)

0.4591*** (0.045)

0.2845*** (0.086) 0.0862* (0.046) -0.2242** (0.090) 0.0662*** (0.017) 0.0673*** (0.021) -0.4727*** (0.092) -0.0497 (0.044) -0.0099 (0.016) -0.8206** (0.352)

0.2263** (0.112) 0.0473* (0.027) -0.1593** (0.075) 0.0662*** (0.020) 0.1160*** (0.029) -0.3774*** (0.104) -0.0480 (0.044) -0.0017 (0.017) -0.8340** (0.417)

YES YES

YES YES

YES YES

573 117 3.326*** 0.049

571 117 6.165*** 0.143

447 93 5.122*** 0.152

SIZE TANGIBILITY PERFORMANCE D_LISTED AGE CONSTANT CONTROL FOR: Time dummies Industry dummies (Pavitt TAxonomy) Observations Number of firms F-statistic R-squared

28

Table 5 – Leverage and diversification: dynamic models

VARIABLES

TOTDEBT_TA (t-1) INT_DIV PROD_DIV INT_PI

FE FE LSDVC (1) (2) (3) TOTDEBT_TA TOTDEBT_TA TOTDEBT_TA

0.2994*** (0.045) 0.4446*** (0.092) 0.2025*** (0.052) -0.4056*** (0.101)

.2757469*** (.0899664) .2999937*** (.1082952) .138712*** (.0628269) -.2750035*** (.1149827) .0770964*** (.022864) .1359025*** (.031887) -.4334519*** (.1119261) -.0397017 (.0463006) -.0285324 (.0183451)

0.2217*** (0.054)

0.2757*** (0.041) 0.3348*** (0.086) 0.1421*** (0.048) -0.2941*** (0.093) 0.0853*** (0.018) 0.1297*** (0.025) -0.4325*** (0.087) -0.0394 (0.039) -0.0211 (0.016) -1.4309*** (0.382)

YES YES

YES YES

YES YES

456 117 11.76*** 0.199

455 117 14.28*** 0.344

338 116

SIZE TANGIBILITY PERFORMANCE D_LISTED AGE Constant CONTROL FOR: Time dummies Industry dummies (Pavitt Taxonomy) Observations Number of firms F-statistic R-squared Sargan test (p-value) AR1 AR2

2.33 ( 0.8011) -3.62( 0.0003) 1.02 ( 0.3089)

29

Table 6 – Leverage and diversification for low and high performance

VARIABLES

(1) (2) (5) (6) TOTDEBT_T TOTDEBT_T TOTDEBT_T TOTDEBT_T A A A A Low High Low High Performance Performance Performance Performance

TOTDEBT_TA (t-1) INT_DIV PROD_DIV INT_PI SIZE TANGIBILITY PERFORMANCE D_LISTED AGE Constant CONTROL FOR: Time dummies Industry dummies (Pavitt Taxonomy) Observations F-statistic R-squared

0.2096** (0.093) 0.0743 (0.049) -0.2104** (0.096) 0.0908*** (0.018) 0.0827*** (0.020) -0.4835*** (0.137) -0.0471 (0.054) -0.0402** (0.019) -1.1882*** (0.384)

0.0060 (0.210) 0.0196 (0.129) -0.1242 (0.220) 0.0538 (0.042) 0.0104 (0.084) -0.5408*** (0.173) -0.0865 (0.088) 0.0198 (0.028) -0.4845 (0.894)

0.2065*** (0.063) 0.2013* (0.113) 0.0661 (0.057) -0.2061* (0.111) 0.0836*** (0.022) 0.1288*** (0.027) -0.3999** (0.160) -0.0171 (0.058) -0.0159 (0.020) -1.2805*** (0.458)

0.2118*** (0.050) -0.1751 (0.169) 0.1455 (0.095) -0.0551 (0.184) 0.0516 (0.043) -0.0503 (0.080) -0.5328*** (0.124) -0.0965* (0.053) -0.0255 (0.021) -0.4051 (0.916)

YES YES

YES YES

YES YES

YES YES

322 88 5.325*** 0.224

249 69 3.696*** 0.108

255 84 5.519*** 0.294

200 67 6.201*** 0.381

30