The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0307-4358.htm

MF
40,3 The capital structure choices
of family firms
Evidence from Italian medium-large
254 unlisted firms
Received 14 March 2013 Pietro Gottardo and Anna Maria Moisello
Revised 10 August 2013 Department of Economics and Business, University of Pavia, Pavia, Italy
18 October 2013
Accepted 20 October 2013
Abstract
Purpose – This paper aims to examine the determinants of capital structure of unlisted firms and
how family governance-related factors impact on them.
Design/methodology/approach – The authors analyze the relation between a set of capital
structure determinants and leverage in a unique dataset of 3,006 family and non-family Italian
medium-large firms (26,210 obs.), and a control sample of 2,730 small firms (14,780 obs.), using
cross-section and panel procedures during 2001-2010.
Findings – Capital structure choices of medium-large family firms are linked to balance-sheet variables
not used in previous studies, i.e. net working capital and capital turnover, and are significantly affected
by the need to maintain control and influence, a relevant dimension of family socioemotional wealth.
Family firms are more levered than non-family firms, but the difference is economically and statistically
significant only for medium-large companies. The presence of the family in active management increases
leverage, as the family endowment in the firm is higher.
Research limitations/implications – This research could be developed through an international
comparison to check the influence of country-related regulatory issues and of national cultural aspects
on family control and influence.
Practical implications – The results can give public authorities important insights in order to
facilitate firms funding specially in the current critical economic scenario and provide managers useful
suggestions to support financial decisions.
Originality/value – To the best of the authors’ knowledge, this is the first paper to explore the
financial choices of a large dataset of medium-large private firms in a bank-based economy.
Keywords Control, Capital structure, Private family firms
Paper type Research paper

1. Introduction
In the last decade numerous studies on family firms’ capital structure choices have been
produced but the factors that drive these decisions are still elusive. Most theoretical and
empirical research focuses on listed companies (Croci et al., 2011; Ampenberger et al.,
2013; Setia-Atmaja et al., 2009; McConaughy et al., 2001; Anderson and Reeb, 2003a, b;
Antoniou et al., 2008; Ellul, 2010; Mishra and McConaughy, 1999; Hagelin et al., 2006;
King and Santor, 2008); relatively few studies analyze private firms and these mostly
concern small-to-medium and small family businesses (Molly et al., 2012; 2010;
Romano et al., 2001; Koropp et al., 2011; Blanco-Mazagatos et al., 2007; Schulze et al.,
Managerial Finance 2001; Lòpez-Gracia and Sànchez-Andujar, 2007; Coleman and Carsky, 1999). This
Vol. 40 No. 3, 2014
pp. 254-275 literature shows that small family firms pursue different financial policies from their
q Emerald Group Publishing Limited non-family peers. Large-scale evidences on private family firms’ capital structure choices
0307-4358
DOI 10.1108/MF-03-2013-0065 are missing (Ampenberger et al., 2012), given the difficulty to obtain reliable data on
private firms. It is therefore desirable to develop studies in this area. We analyze a large Capital structure
dataset of Italian private family and non-family firms over the period 2001-2010 focusing choices of family
on medium-large firms, for which ownership and management information is available.
To the best of our knowledge this paper is the first to analyze financing decisions of firms
medium-large private family firms in a bank-oriented economy. The capital structure,
as argued by Antoniou et al. (2008) is heavily influenced by the environment and
institutions, corporate governance and tax systems, capital market development and 255
borrower-lender relationship. Italy has an underdeveloped stock market compared to
Anglo-Saxon countries, with only 271 listed firms including banks, insurances and
other financial holdings, a market capitalization that accounts just for 27 percent of
GDP (2010), and a bank-oriented financial system with widespread relationship
lending. Medium-large firms cover 51 percent of the Italian industrial firms turnover.
Private firms account for 83 percent of the industrial value-added (2010), weighting
much more heavily in the Italian economy with respect to peer countries like Germany
and France. We share with these countries a very similar industrial structure, while the
main differences are a larger proportion of small firms and a greater relevance of
unlisted firms. Family firms represent 69 percent of the population of Italian
medium-large private firms, confirming the prominent role of families in the Italian
industrial structure.
We find that medium-large private family and non-family firms show differences in
capital structures analogous to those found in previous studies for listed firms. Our
empirical evidence indicates that family and non-family firms present different
characteristics in terms of leverage, size, firm market share, profitability, ownership
concentration and ownership involvement in active management; such differences are
significant also at the industry level. The regression analyses, show how capital
structure determinants affect leverage in different ways, depending on whether a
company is family controlled or not. Our findings point out that family firms financing
decisions do not depend only on economic and financial factors but are significantly
affected by ownership and governance-related factors, that is the need to maintain the
family endowment in the firm retaining control and influence. We find that family firms
have higher leverage ratios than non-family firms, but the differences are economically
and statistically significant only for medium-large companies. Moreover, the family
presence in active management leads to significantly higher levels of leverage, as the
family endowment in the firm, in terms of degree of family identification, influence and
personal investment in the firm is higher. The reminder of the paper is organized as
follows: Section 2 reviews some of the more recent papers focusing on the characteristics
of family firms; Section 3 reviews the capital structure literature on family firms;
Section 4 describes dataset, sample selection criteria and variables; Section 5 provides
descriptive statistics; Section 6 shows the correlation, the regression analysis and the
discussion of the results; Section 7 concludes and provides avenues for future research.

2. The theory and the empirical literature on family firms

Principal-agent theory assumes that the separation of ownership and control
(Fama and Jensen, 1983) causes conflicts of interest and asymmetric information
between the owner and whoever exercises control. The latter (agent) tends to behave in
his or her own interest rather than the interest of the former (principal), giving rise to a
moral hazard (Hölmstrom, 1979) and adverse selection ( Jensen and Meckling, 1976).
MF The personal ownership involvement discourages managers (agent) from acting
40,3 against ownership (principal) interest through the misallocation of resources
(Schulze et al., 2002). As a result, Daily and Dollinger (1992) claim that family firms
need comparatively lower costs for their control, but Schulze et al. (2001) observe that
different types of agency cost weigh on family firms, related to altruistic behavior of
the owner-manager towards employed family members, to management entrenchment
256 and to shareholder expropriation.
Gomez-Mejia et al. (2007) developed a general “socioemotional wealth model” to
explain family firms peculiarities. The model is based on the behavioral agency theory
(Wiseman and Gomez-Mejia, 1998; Gomez-Mejia et al., 2000), which integrates elements
of prospect theory, behavioral theory of the firm and agency theory. This theory
focuses on the concept that firms decisions depend on the perspective of the firm’s
dominant principals, which are concerned with the preservation of the endowment
built up in the firm. Fundamental to family firm’s principals is the conservation of the
socioemotional endowment, which refers to “the stock of affect related value that a
family derives from its controlling position in a particular firm” (Berrone et al., 2012).
During the last decade the financial economic literature has shown an increasing
interest in different aspects of family firms. A first stream of studies focused on the
comparison between family and non-family firms’ performance, leading to mixed
results (Anderson and Reeb, 2003a, b; Villalonga and Amit, 2006; Barontini and
Caprio, 2006; Sacristán-Navarro et al., 2011). The studies on investment choices pointed
out that many family firms follow long-term investment strategies in order to
perpetuate their business for future family members (Gallo and Vilaseca, 1996;
McConaughy and Philipps, 1999) and they pursue business stability rather than rapid
growth and risky investment strategies (Harris et al., 1994; Tagiuri and Davis, 1992).
Other research has studied family influence on firms’ strategic behavior in
general (Miller et al., 2011), or has focused on dividend policy (Gugler, 2003 or
Correia da Silva et al., 2004), or on diversification decisions (Anderson and Reeb, 2003b
or Gomez-Mejia et al., 2010).
There is some evidence that family firms’ corporate decision making pursues very
conservative strategies (Klasa, 2007; Bauguess and Stegemoller, 2008; Sraer and
Thesmar, 2007; Caprio and Croci, 2011) but not much is known about the nature of
their capital structure decisions or whether they differ from those of non-family firms.

3. The capital structure of family firms

An important part of the research studied the choices of family firms capital structure
by testing the main theories on firms’ capital structure decisions, pecking order theory
(POT) and trade-off theory (TOT). According to POT, firms do not pursue any leverage
ratio target and they meet their capital needs via the following order of preference:
accumulated earnings, short-term borrowing, long-term borrowing and equity
issuance (Donaldson, 1961; Myers and Majluf, 1984). TOT, on the other hand,
assumes that firms have a leverage ratio target determined by balancing the costs and
benefits of debt and equity (Titman, 1984). The international empirical evidence gives
mixed results but the POT appears to fit better firm behavior, both in family and
non-family businesses (Rajan and Zingales, 1995; Shyam-Sunder and Myers, 1999;
Lòpez-Gracia and Sànchez-Andujar, 2007; Frank and Goyal, 2009; Sheikh and Wang,
2011; Graham and Leary, 2011).
 Mishra and McConaughy (1999) point out that US family firms present a Capital structure
significantly lower level of debt. Further, they show that this is related to the choices of family
peculiarities of the founding family rather than to the level of managerial ownership.
Anderson and Reeb (2003b) find that US family firms have a higher level of debt than firms
their non-family counterparts and Ellul (2010), analyzing an international dataset of
companies from 36 countries, confirms these results. King and Santor (2008) and
Setia-Atmaja et al. (2009) find higher debt ratios in Canada and Australia. In Europe, 257
Ampenberger et al. (2013), studying a dataset of listed German companies, and
Margaritis and Psillaki (2010), focusing on French firms from the manufacturing
industry, find the opposite result. Hagelin et al. (2006) find that, on average, Swedish
family-controlled firms do not rely on less debt than non-family firms do. Croci et al.
(2011), for a European dataset, come to a different conclusion in finding that family
firms have a preference for debt financing and a strong aversion to equity financing
consistent with the value-of-control hypothesis. This argument is also supported by the
studies conducted by Brav (2009) who shows that private UK firms, unlike public ones,
rely almost exclusively on debt financing and concludes that control motives affect the
choice of external financing. Furthermore, founding families view the firm as an asset
that will be handed down to future generations (Chami, 1999) and this consciousness
enhances the control motivation. Family firms’ high debt ratios are also related, as
Croci et al. (2011) argue, to their cost of debt (lower than for non-family firms) and cost
of equity (higher than for non-family firms). Moreover, long-term orientation facilitates
connections and relationships between family businesses and stakeholders, among
them, external financiers (Carney, 2005; Haynes et al., 1999). Anderson et al. (2003) find
that family ownership reduces the cost of debt financing as a result of lower agency
conflicts between equity and debt holders. This means bondholders perceive family
ownership as a corporate structure that protects their interests well. In line with these
results, Croci et al. (2011) show that family firms have a preference for low-risk
investments and that credit markets provide more long-term debt to family firms,
indicating that markets view their investment decisions as less risky and that the
appeal of debt financing in family firms is related to the adverse selection costs of
equity. This is caused by information asymmetries related to agency problems, which
might develop vertically between owners and managers, as well as horizontally
between owners and family firm shareholders (Villalonga and Amit, 2006).
Since family blockholders tend to have low diversification (Anderson and Reeb,
2003b), family firms have to manage the trade off between increasing leverage to retain
control and reducing leverage in order to minimize the firm’s risk.
Molly et al. (2010), for a sample of small-medium businesses point out the effect of
succession on leverage and show that a transfer from the first to the second generation
negatively influences the debt ratio of the firm, whereas in successions between later
generations this effect is reversed. Moreover, Molly et al. (2012) investigate the
intergenerational effect on growth and corporate structure finding that next-generation
firms grow slower because they prefer to forego part of their growth rather than risk
the loss of family control due to the increased use of debt.
González et al. (2013), by studying a sample of private and listed firms belonging to
Columbian business groups stress the tradeoff between two distinct motivations that
determine the capital structure of family firms: risk aversion pushes firms toward
MF lower debt levels, but the need to finance growth without losing control makes family
40,3 firms to prefer higher debt levels.
According to the logic of the socioemotional wealth we expect that family firms
would be prone to increase leverage in order to maintain control and family influence,
as it is a fundamental dimension of the socioemotional endowment (Gomez-Mejia et al.,
2007). Socioemotional wealth takes priority over risk aversion (Berrone et al., 2012). On
258 these bases we hypothesize family businesses to be more levered than non-family firms
and we expect this feature to be particularly relevant in the case of medium-large firms,
which have already faced the problem of how to finance the growth and manage the
trade-off between equity and debt.
The empirical findings of Gomez-Mejia et al. (2007), show that a stronger role of the
family makes more likely the firm to strive to protect its socioemotional wealth. Families
can actively exert control over strategic firm decisions appointing a family member as
CEO and/or chairman of the board, Family identification, influence and personal
investment in the firm increase, so we expect the firms with family members involved in
active management to be more levered than family firms professionally managed.

4. Data and variable definition

We analyze a unique dataset extracted from AIDA (Italian Digital Database of
Companies), which is the Italian provider of the Bureau Van Dijk European Database,
and which constitutes the most complete and reliable financial information source
about non-public companies. AIDA database contains detailed accounts following the
scheme of the 4th EU Directive for up to ten years, shareholders, anagraphic and
industrial sector information and, in the case of medium-large firms, CEO and board
composition. We selected non-financial firms with the following characteristics:
a registered company office in Italy, active in the year 2010 in the form of limited
company (Società a Responsabilità Limitata – SRL), or company limited by shares
(Società per Azioni – SPA), and revenues of over e70,000,000 in one or more years from
2001 to 2010 inclusive; in total we isolated 3,006 firms. We chose only non-financial
companies to avoid the effect of financial sector regulation and peculiarities on firms’
financing decisions and we established the minimum revenues to ensure the
availability of the basic data for our analysis. We hand-checked all balance-sheet data
from the period 2001-2003 in order to correct any errors resulting from the transition
from Lira to Euro. We also checked all ratios for outliers to avoid them distorting the
analysis results. We completed the database by entering the data on ownership and
governance that were not in the AIDA database, taking them from the Italian Chamber
of Commerce Register.
We define a firm as being a family firm when the founder, or descendents of his or
her family (either by blood or through marriage), is a blockholder, individually or as a
group, and we assume an ownership cut-off point of 20 percent. Previous research has
used different cut-off points but our choice is consistent with a large body of literature:
La Porta et al. (1999), Faccio and Lang (2002), Dahya et al. (2008), Ellul (2010) and
Croci et al. (2011). La Porta et al. (1999) and Faccio and Lang (2002) also use 20 percent
as a threshold, other authors employ 20 and 25 percent (Andres, 2008; Franks et al.,
2010). Consequently, our database of medium-large firms is made up of 2,077 family
and 929 non-family firms. In the regression analysis, in order to better check
our hypothesis, we use two family-firm definitions: the first is ownership-based,
as previously specified, the second based on the presence (family firms) or non Capital structure
presence (non-family firms) of family members in active management, without an choices of family
ownership threshold.
To better highlight the peculiarities of medium-large family firms we also analyze a firms
control dataset of small businesses, extracting from AIDA a random sample of 3,000
non-financial firms with revenues in the range e5-10 millions. On the basis of the available
shareholders information we retained 2,730 firms: 2,082 family and 648 non-family 259
businesses. The AIDA database does not contain information on CEO and board
composition for this size class. To define a family firm we apply the above classification
criterion assuming an ownership cut off point of 50 percent. This choice is consistent with
previous studies on small family firms (Molly et al., 2012; Blanco-Mazagatos et al., 2007;
Lòpez-Gracia and Sànchez-Andujar, 2007; Coleman and Carsky, 1999).
In our descriptive statistics, correlation analysis and regression models, we use a
definition of leverage corresponding to book leverage, as it is widely used in existing
literature:
Longterm financial debts þ Financial debt in current liabilities
Book leverageðBLevÞ ¼
Total assets
Various firm characteristics have been found to affect capital structure decisions
(Rajan and Zingales, 1995; Lemmon and Zender, 2010; Ellul, 2010; Marchica and Mura,
2010), especially profitability, asset tangibility, size, liquidity and depreciation.
We measure profitability with the ratio of operating income to total assets;
tangibility is the ratio of net property, plant and equipment to total assets. We use two
measures of size: the natural logarithm of sales – Log(Sales) – and the total assets
value. Liquidity is determined by the ratio of liquid assets to total assets; depreciation
is the ratio of depreciation to total assets.
To these variables, used in previous research, we add: firm’s market share, cash
flow, net working capital, inventory, capital turnover. To measure a firm’s market
share we match firms to industries requiring a non-missing ATECO code in the
database and we determine market shares on firms’ revenue basis. In doing this we use
all available firms with revenues of over e70,000,000 as of 2010 and exclude firms with
missing revenues. Cash flow, is measured with the ratio of operational cash flow to
total assets; net working capital is computed as the difference between current assets
and current liabilities related to total assets; inventory is the ratio of its value to current
assets; and capital turnover is the ratio of net sales to capital employed.
In order to describe and analyze firms’ governance features we use family
ownership, which measures the percentage of equity owned by the family, and
management ownership as the percentage of equity owned by a firm’s management.
Moreover, we use a dummy variable, firms foreign owned, which indicates if the firm is
under control of a foreign firm.

5. Descriptive statistics
Table I provides descriptive statistics for firm-level financial and ownership
characteristics for medium-large and small firms samples, for the family and
non-family businesses. The last column shows the results of the t-test, which verifies
the significance of differences of the two samples’ mean values. The medium-large
firms descriptive statistics show a highly significant difference in the financial
 MF
40,3

260

Table I.
Descriptive statistics
Medium-large firms Small firms
Non-family Non-family
All firms Family firms firms All firms Family firms firms
Mean Median Mean Median Mean Median t-test Mean Median Mean Median Mean Median t-test

Book leverage 23.71 22.10 26.18 25.52 18.11 14.12 211.50 * * 18.83 14.71 19.06 14.98 18.11 14.08 2 1.15
Log (sales) 11.58 11.39 11.44 11.31 11.90 11.64 10.90 * * 8.90 8.90 8.89 8.89 8.90 8.90 1.50
Firm market share 1.83 1.26 1.74 1.20 2.01 1.31 2.94 * * 0.10 0.06 0.09 0.06 0.10 0.06 2.85 * *
Profitability 4.74 4.06 5.08 4.23 4.14 3.53 2 3.20 * * 5.80 4.58 5.58 4.46 6.51 4.98 1.14
Cash flow 5.49 4.94 5.91 5.08 4.68 4.58 2 1.16 5.78 4.73 5.58 4.67 6.41 4.96 3.17 * *
Depreciation 3.67 3.00 3.52 2.95 4.02 3.21 3.77 * * 3.53 2.89 3.45 2.87 3.79 2.95 2.45 *
Tangibility 16.88 12.59 17.25 14.17 16.03 9.20 2 1.74 21.24 16.01 21.41 16.88 20.69 12.28 2 0.79
Total assets 280.96 75.26 158.35 66.79 559.69 104.58 3.59 * * 6.72 5.45 6.34 5.50 7.95 5.22 2.68 * *
Net working capital 3.06 3.59 2.77 2.74 4.19 5.86 0.92 17.64 15.66 17.77 15.95 17.23 14.80 2 0.56
Liquidity 5.55 3.14 5.59 3.27 4.40 2.73 2 0.43 8.20 4.21 8.21 4.09 8.16 4.41 2 0.10
Inventory 26.22 23.72 28.99 26.56 20.07 17.02 211.45 * * 26.04 22.74 27.98 24.95 19.83 15.28 2 9.49 * *
Capital turnover 1.45 1.23 1.47 1.24 1.41 1.21 2 1.71 1.52 1.33 1.52 1.31 1.55 1.38 0.75
Family ownership 63.36 95.36 90.99 100 12.46 0 72.95 100 95.65 100 0 0
Management ownership 39.01 16.65 54.65 60 3.35 0 – – – – – –
NoEx board ownership 12.18 0 16.82 0 1.62 0 – – – – – –
Firms foreign owned 24.44 0 9.34 0 59.01 100 2.97 1.15 8.80
Cost of debt 7.10 6.29 6.87 6.15 7.79 6.83 2.58 * * 5.67 5.54 5.66 5.57 5.70 5.51 0.26
Obs. 3,006 2,077 929 2,730 2,082 648
Notes: Significant at: *5 and * *1 percent levels; the table reports the mean and median of various firm characteristics for family and non-family firm’s
sub-samples and separately for medium-large and small firms; the medium-large firms sample consist of all non-financial firms in the BVD Aida database,
covering the time span 2001-2010, extant in 2010, with sales . 70 ml e, the small firms sample represents firms with sales in the range 5-10 ml e; book
leverage is the ratio of total financial debt to total assets; firm size is measured as the log of sales, we report also total assets in ml e; market share is the
ratio of 2010 firm revenues to industry total revenues; profitability is the ratio of operating income before depreciation to total assets; cash flow, is the ratio
of operational cash flow to total assets; depreciation is the ratio of depreciation of physical and intangible assets to total assets; tangibility is the ratio of
net property, plant and equipment to total assets; net working capital is the ratio of current assets minus current liabilities to total assets; liquidity is the
ratio of liquid assets to total assets; inventory is the ratio of inventory to current assets; family ownership is the sum of the voting rights of all family
members; management ownership is the sum of direct and indirect voting rights in % owned by all top executives; firms foreign owned indicates to what
extent the firms in the sample are controlled by a foreign firm capital turnover is the ratio of sales to capital employed; tests of the differences in means
between family and non-family firms are displayed for the medium-large and small firm’s samples
structure of family and non-family firms. The data suggest an higher book leverage for Capital structure
family firms, with a mean of 26.18. For non-family firms, the mean book leverage is choices of family
18.11; for all firms the mean value is 23.71. This result shows a significant difference in
the financial choices of private family and non-family firms and confirms the result of firms
empirical studies carried out by Anderson and Reeb (2003b), Ellul (2010), King and
Santor (2008) and Setia-Atmaja et al. (2009) showing that firms with entrenched
managers select less risky investments and use more debt finance; it is also consistent 261
with theoretical papers (Fulghieri and Suominen, 2012) that argue that poor corporate
governance (entrenched firms) can lead to greater debt financing.
Both the size measures adopted, sales and total assets, present lower values for the
family firms. They present a sales mean of 11.44 while, for non-family firms, the figure
is 11.90; for the entire panel it is 11.58. The total assets mean is e158.35 million for
family firms, e559.69 million for non-family firms and e280.96 million for all firms.
These results, statistically significant at the 1 percent level, taking into account the
higher book leverage, may indicate that family firms are growing with an annual
median growth rate of 8.4 percent (4.5 percent for non-family firms) but this growth is
conditioned by the need to retain control. The market share of family firms is, on
average, lower (1.74) than non-family firms’ (2.01) although the median values are quite
close: 1.20 for family and 1.31 for non-family firms. Even if family firms present, on
average, a lower firm market share compared to the non-family control sample, they
have a higher profitability level: the family mean value is 5.08 and non-family is 4.14.
The higher level of profitability could be due to a higher capital turnover ratio: 1.47 for
family firms and 1.41 for non-family ones, but this difference in the means is not
statistically significant. The results, in terms of profitability, are in contrast with the
evidence described by Ellul (2010), on an international dataset of listed companies, but
are in line with those of Barontini and Caprio (2006) on a dataset of publicly-traded
corporations from 11 countries in continental Europe. Andres (2008) finds, for a dataset
of German companies, that performance is higher for family firms than non-family
when the family has board representation. Anderson et al. (2003) observe that, when
family members serve as CEOs, performance is better than with outside CEOs. The
high degree of family involvement in active management in our dataset is likely to
positively affect profitability levels. As we can see, family firms’ ownership has a high
degree of involvement in the management as executive board ownership mean value is
54.65, while non-executive board member ownership mean value is 16.82. On the
contrary, non-family firms present a very low degree of management entrenchment as
executive board ownership mean value is 3.35 versus a non-executive board ownership
of 1.62 and more than half of these firms do not have owners involved in the
management or on the non-executive board. Family ownership is very concentrated; in
fact we observe a mean value of 90.99. Family firms, unlike non-family ones, are
characterized by a strictly national ownership: only 9.34 percent are foreign owned
(versus a value of 59.01 percent for non-family firms).
From the assets characteristics we observe that tangibility shows a higher mean
value for family (17.25) than for the non-family firms (16.03) (Anderson et al., 2010). We
observe lower levels of depreciation for family (3.52) than for non-family firms (4.02).
This effect is likely to be due to the weight of land and buildings, not subject to
depreciation, on the aggregate fixed assets or to a conservative plant and machinery
management with longer depreciation periods. However, we do not have the fixed asset
MF data in disaggregated form to check these hypotheses. A higher liquidity ratio
40,3 characterizes the family firms, which reach 5.59 versus a value of 4.40 for non-family,
but, for the former, a higher level of current debt limits the ability to cope with
monetary liabilities, as the net working capital is 2.77 for family firms and 4.19 for
non-family. Moreover, inventory presents a higher weight on working capital for
family firms, as the mean value is 28.99 versus a value of 20.07 for non-family ones.
262 This result could be due to basic inventory control systems related to the conservative
behavior of family firms and the high level of family involvement in management: on
the one hand it could imply the employment of non-professional family members or
make the company less attractive for professional, highly-qualified managers,
as Schulze et al. (2001) study demonstrate.
Coherent with Anderson et al. (2003), the cost of debt is lower in family firms (with a
mean of 6.87) than in non-family ones (with a mean of 7.79), as credit markets perceive
the former less risky than the latter (Croci et al., 2011).
These statistics indicate that medium-large family and non-family firms have
different characteristics both from the economic and financial magnitudes point of
view and from that of governance and that they are statistically significant at the
1 percent level for financial structure (leverage), profitability, size (log(sales) and total
assets), firm market share, inventory, depreciation and cost of debt.
Analyzing the small firms statistics we see that the main differences between family
and non-family firms tend to persist. Nevertheless, small family businesses present a
different balance sheet structure in terms of higher liquidity, tangibility and net
working capital. As we expected, we observe that small family are more levered than
non-family ones, even if the difference is not statistically significant – this result is in
line with the findings of Coleman and Carsky (1999) – but at the same time we see that
small firms are much less levered than their medium-large counterparts. Small firms
are almost all domestic regardless of family ownership.
The characteristics of medium-large firms could be related to industry peculiarities
and the results could be affected by the distribution across industries of family and
non-family firms’ and industries weight in the database. As Cohen and Yagil (2010)
suggest, industries may have specific financial needs caused by different operating
conditions and also an imitation effect may occur within them; so, in Table II, we show
the descriptive statistics computing the variables’ mean value for each industry. As
Italian ATECO codes up to four digits are coherent with European standards, we base
our classification on four-digit ATECO codes and we assign firms to 27 industries.
Single-industry family firms, on average, tend to maintain the characteristics observed
in the entire dataset. Family firms presents a higher book leverage in 85 percent of
industries, they are smaller in 92 percent, more profitable in 69 percent, have a higher
level of depreciation in 69 percent and a higher inventory in 77 percent. Family firms’
net working capital mean value is higher in 42 percent of industries but the median
value confirms the tendency observed at the entire dataset level, in fact family firms’
net working capital median value is higher in 65 percent of industries.

6. The determinants of family firm capital structure. Correlation,

regression analysis and discussion of the results
Table III summarizes the correlation analysis results. Book leverage shows a
significant relation with both economic-financial and governance related factors.
 Family firms Non-family firms
Sector BLev LnS Prof Depr NWC Inv n BLev LnS Prof Depr NWC Inv n

Apparel 19.08 11.64 8.18 3.90 18.01 29.30 51 22.30 11.75 0.46 3.05 5.08 23.68 5
Food 35.7 11.48 4.47 3.62 23.15 29.97 153 24.96 11.88 21.09 6.02 21.40 23.45 19
Paper 31.80 11.73 3.20 5.21 25.37 25.15 22 15.10 11.96 4.54 5.46 7.16 18.78 10
Chemicals 24.88 11.52 4.86 4.36 2.47 22.39 73 19.20 11.71 5.07 5.22 6.65 21.74 57
Wholesale and retail trade 26.27 11.45 5.48 3.08 2.67 26.33 526 18.08 11.79 5.05 2.84 8.16 19.88 228
Construction 21.13 11.00 3.08 1.59 1.54 47.13 131 19.51 11.53 3.85 2.66 23.18 21.88 42
Printing, publishing 19.99 11.64 3.66 5.07 23.92 10.79 23 20.98 12.22 4.90 3.62 3.92 12.19 12
Computer and electronics 25.18 11.25 5.26 3.82 5.85 21.77 24 16.71 11.84 3.86 5.62 12.34 24.77 32
Household and other elect. 21.10 11.51 7.33 3.61 12.27 26.00 53 18.80 11.88 4.57 3.70 10.98 24.71 42
Coal, oil and gas 25.86 12.57 5.71 4.28 5.75 17.62 13 28.92 13.42 1.55 5.04 29.45 23.40 13
Farmaceutical 16.64 11.97 9.30 4.83 14.72 26.21 27 23.65 12.12 8.36 4.77 6.33 25.81 29
Beverage 24.30 11.40 9.48 3.72 0.19 22.98 18 30.31 12.05 1.97 4.92 214.22 20.67 10
Rubber 27.40 11.42 5.17 5.54 0.10 23.34 41 24.12 11.71 4.56 5.56 25.10 21.62 26
Commun., web and data processing 19.89 11.38 3.53 4.02 29.17 5.83 18 11.01 12.00 4.46 7.43 6.19 4.35 33
Building materials and glass 22.42 11.46 5.46 4.73 4.96 28.02 56 13.34 11.78 6.85 5.12 3.98 28.97 15
Machinery and computer equip. 21.27 11.40 6.13 3.45 9.79 31.99 140 16.59 11.64 4.00 3.81 7.02 31.83 72
Metal industries 28.46 11.44 5.40 4.26 6.26 32.28 225 19.89 11.66 2.09 4.10 4.66 32.72 42
Wood and furniture 30.80 11.30 3.68 3.64 22.18 32.03 38 5.53 11.35 5.83 2.61 19.16 18.49 3
Transport. equip. and related services 27.99 11.51 3.41 2.60 3.75 49.11 173 18.49 12.32 2.00 4.16 2.58 31.48 59
Leather and leather product 32.96 11.49 6.40 3.17 8.80 35.94 31 26.98 11.19 4.39 4.37 21.43 20.95 4
Motion, theatre and entert. 17.20 11.82 0.77 10.64 214.15 1.87 9 15.58 12.06 22.29 16.08 217.15 0.50 5
Lodging and eating drinking services 22.32 11.28 4.82 3.54 216.63 8.67 18 11.04 11.70 3.87 4.79 23.58 4.12 9
Utilities 24.75 11.37 4.29 2.69 26.17 10.30 55 17.45 12.28 3.00 3.25 26.99 4.55 93
Textile 27.73 11.43 4.00 4.89 10.97 35.37 36 – – – – – – 0
Transportation and related services 27.39 11.45 4.97 3.94 210.99 5.31 98 13.89 12.06 2.82 3.56 5.64 4.03 53
Miscellaneous manufact. 31.43 11.45 5.31 2.63 8.84 28.01 16 31.05 11.51 2.36 4.06 21.85 33.52 4
Other business services 19.66 11.57 2.86 3.49 20.92 29.07 9 7.80 11.55 7.11 2.58 25.34 26.97 12
Notes: The table reports the industrial distribution and the mean of various firm characteristics for medium-large family and non-family firm’s sub-samples; the
sample consist of all non-financial firms in the BVD Aida database, covering the time span 2001-2010, extant in 2010, with sales . 70 ml e; book leverage (BLev) is
the ratio of total financial debt to total assets; firm size (LnS) is measured as the log of sales; profitability (Prof) is the ratio of operating income before depreciation to
total assets; Depr is the ratio of depreciation of physical and intangible assets to total assets; net working capital (NWC) is the ratio of current assets minus current
liabilities to total assets; inventory (Inv) is the ratio of inventory to current assets; n represents the number of family or non-family firms in each of the 27 two-digit
industries listed in the table
Capital structure

firms
choices of family

medium-large firms
Industrial distribution of
263

Table II.
 MF
40,3

264

Pearson’s
Table III.

correlation matrix,
medium-large firms
LnS FMS Prof Depr Tang NWC Liq Inv CTR FOwn MOwn

BLev 2 0.078 2 0.126 20.176 20.087 0.133 20.462 20.284 0.100 2 0.071 0.174 0.156
(0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001)
LnS 0.078 0.013 0.076 0.008 20.041 20.097 20.086 0.078 2 0.227 2 0.198
(0.001) (0.48) (0.001) (0.64) (0.025) (0.001) (0.001) (0.001) (0.001) (0.001)
FMS 0.028 0.144 0.024 20.021 0.046 20.069 2 0.003 2 0.054 2 0.073
(0.12) (0.001) (0.18) (0.25) (0.012) (0.001) (0.88) (0.003) (0.001)
Prof 20.084 20.011 0.344 0.178 20.037 0.144 0.058 0.084
(0.001) (0.56) (0.001) (0.001) (0.040) (0.001) (0.002) (0.001)
Depr 0.401 20.154 20.036 20.104 2 0.179 2 0.055 2 0.068
(0.001) (0.001) (0.048) (0.001) (0.001) (0.003) (0.001)
Tang 20.359 20.140 20.020 2 0.241 0.051 0.026
(0.001) (0.001) (0.26) (0.001) (0.006) (0.15)
NWC 0.222 0.017 0.165 2 0.013 0.017
(0.001) (0.34) (0.001) (0.49) (0.34)
Liq 20.118 0.125 0.017 0.041
(0.001) (0.001) (0.36) (0.026)
Inv 2 0.070 0.185 0.171
(0.001) (0.001) (0.001)
CTR 0.042 0.041
(0.022) (0.026)
Notes: The table reports Pearson correlations and p-values for the full sample of medium-large firms; the sample consist of all non-financial firms in the
BVD Aida database, covering the time span 2001-2010, extant in 2010, with sales . 70 ml e, it comprises 3,006 firms; book leverage (BLev) is the ratio of
total financial debt to total assets; firm size is measured as the log of sales (LnS); market share (FMS) is the ratio of firm revenues to industry total
revenues; profitability (Prof) is the ratio of operating income before depreciation to total assets; depreciation (Depr) is the ratio of depreciation of physical
and intangible assets to total assets; tangibility (Tang) is the ratio of net property, plant and equipment to total assets; net working capital (NWC) is the
ratio of current assets minus current liabilities to total assets; liquidity (Liq) is the ratio of liquid assets to total assets; inventory (Inv) is the ratio of
inventory to current assets; capital turnover (CTR) is the ratio of sales to capital employed; family ownership (FOwn) is the sum of the voting rights of all
family members; management ownership (MOwn) is the sum of direct and indirect voting rights in % owned by all top executives; p-value in parentheses
Book leverage is negatively and significantly related with size, firm market share, Capital structure
profitability, depreciation, net working capital, liquidity, and capital turnover ratio and choices of family
is positively related to tangibility, inventory, family ownership and management
ownership. This is to say that, on the one hand, as size, the ability to meet monetary firms
commitments and generate internal resources for financing increase, leverage
decreases. On the other hand, as inventory grows, short-term debt and, consequently,
leverage increases. The findings of the correlation analysis conducted on the entire 265
dataset are coherent with POT predictions (Romano et al., 2001).
Focusing on family firms, we observe that the correlation analysis on governance
related factors shows that leverage tends to increase as family ownership grows and
this result is consistent with González et al. (2012) findings for a sample of large firms.
Moreover, we observe that, consistently with King and Santor (2008), family ownership
is positively related with leverage and profitability consistently with the descriptive
statistics which show that family firms are more profitable and, as we expected, more
levered than non-family ones. Family firms’ financing decisions cannot be explained
purely by POT because they are also affected by governance-related motivations
(Romano et al., 2001). These findings also confirm those of Ellul (2010), who interprets
family firms’ higher leverage as a means, for the controlling family, to maintain control
of the firm’s management and those of Gomez-Mejia et al. (2007) which show that
family firms exhibit a stronger preference to retain owner control of the organization
than non-family firms.
To test our hypothesis we carry out a cross-sectional analysis. We run a model
specification for family firms selected on an ownership basis, family firms identified
according to family members’ presence in active management, non-family firms
defined through ownership criterion and non-family firms identified as those with no
family members in active management. In the regression analysis we use variables
commonly used in previous research on firms’ capital structure, that is their size,
growth, profitability, tangibility and liquidity. As these variables explain only a small
fraction of leverage variability, we add inventory, capital turnover, family in active
management, and net working capital.
Table IV shows the cross-sectional analysis results with heteroskedasticity robust
standard errors as the White test reported in the table always rejects the hypothesis of
homoskedasticity of the residuals. Consistent with Rajan and Zingales’ (1995)results on
German listed companies, we observe that size, for medium-large non-family firms
have a significant negative relation with leverage, for family firms the negative
coefficient is not significant. Conversely, for small family firms we find a positive
relation, significant only for the family firms (Blanco-Mazagatos et al., 2007;
Lòpez-Gracia and Sànchez-Andujar, 2007; Gallo and Villaseca, 1996; Romano et al.,
2001). The relation between profitability and leverage is positive for medium-large
firms, as for Croci et al. (2011) sample of listed firms, but never significant. The
coefficient is negative and significant only when the model specification does not
include net working capital[1]. On the contrary small family firms present a significant
negative relationship, confirming the results of Lòpez-Gracia and Sànchez-Andujar
(2007) and Molly et al. (2010), respectively, for small and small to medium firms.
A firm’s market share also presents a negative relation with leverage and it is
stronger in family firms, but this results holds only for medium-large firms. For family
firms tangibility is negatively related with leverage and the significant negative
 MF
40,3

266

Table IV.

capital structure
Cross-sectional analysis
of family ownership and
Medium-large firms
Family firms Non-family firms Small firms
Model Own $ 20 FMngt¼ 1 Own , 20 FMngt¼ 0 Family firms Non-family firms

Interc 0.391 (6.31) 0.299 (4.38) 0.400 (6.72) 0.419 (8.61) 2 0.981 (24.01) 20.419 (2 0.96)
LnS 20.009 (2 1.64) 0.004 (0.61) 2 0.016 (23.28) 20.016 (2 4.12) 0.153 (5.59) 0.085 (1.74)
FMS 21.136 (2 7.05) 21.101 (25.35) 2 0.343 (21.35) 20.747 (2 4.14) 2.725 (0.84) 25.215 (2 1.04)
Prof 0.030 (0.64) 0.052 (0.74) 0.024 (0.26) 0.062 (1.01) 2 0.153 (22.49) 0.083 (1.40)
Tang 20.072 (2 3.00) 20.081 (23.00) 2 0.047 (21.12) 20.030 (2 0.92) 2 0.078 (23.28) 20.025 (2 0.54)
Liq 20.638 (2 11.21) 20.799 (211.98) 2 0.281 (25.09) 20.245 (2 4.49) 2 0.314 (28.73) 20.321 (2 5.07)
Inv 0.005 (0.42) 20.003 (20.27) 0.103 (3.07) 0.098 (3.96) 0.143 (8.18) 0.159 (4.37)
CTR 0.006 (1.55) 0.013 (2.83) 2 0.008 (21.18) 20.005 (2 1.04) 2 0.086 (216.98) 20.064 (2 8.51)
FMngt 0.048 (6.16) 0.060 (1.55)
NWC 20.361 (2 21.48) 20.337 (216.51) 2 0.301 (24.78) 20.332 (2 6.82) 2 0.292 (213.59) 20.329 (2 9.49)
Obs. 2,053 1,521 920 1,484 2,082 648
White test 129.40 88.73 221.51 257.20 108.51 38.56
Adj. R 2 0.31 0.32 0.27 0.28 0.37 0.34
Notes: The medium-large firms sample consist of all non-financial firms in the BVD Aida database, covering the time span 2001-2010, extant in 2010,
with sales . 70 ml e, e, the small firms sample represents firms with sales in the range 5-10 ml e; the table presents parameter estimates from cross-
sectional regressions of book leverage; the t-statistics in parentheses are computed using standard errors robust for heteroskedasticity, as the White test
statistics shows that the null hypothesis of homoskedasticity is rejected at the standard 1 percent level; family firms are defined in two ways: (1) family-
owned firms in which one or more family members are individually or as a group blockholders, with a minimum threshold ownership of 20 percent; (2)
family controlled and managed firms whose CEO and eventually some other executive director is a member of the blockholding family; book leverage
(BLev) is the ratio of total financial debt to total assets; firm size is measured as the log of sales (LnS); market share (FMS) is the ratio of firm revenues to
industry total revenues; profitability (Prof) is the ratio of operating income before depreciation to total assets; tangibility (Tang) is the ratio of net property,
plant and equipment to total assets; liquidity (Liq) is the ratio of liquid assets to total assets; inventory (Inv) is the ratio of inventory to current assets;
capital turnover (CTR) is the ratio of sales to capital employed; FMngt is a dummy that in family firms takes value one if one or more shareholders in active
management are family members and zero otherwise, in non-family firms it takes value one if a shareholder is in active management and zero otherwise;
net working capital (NWC) is the ratio of current assets minus current liabilities to total assets; t statistics corrected for heteroskedasticity in parentheses
correlation between tangibility and net working capital accounts for this effect. Capital structure
Unsurprisingly we find a negative relation with liquidity (Anderson and Reeb, 2003b). choices of family
Inventory has some explanatory power only in the case of non-family firms and for
small family firms for which the growth of the inventory results in an increase in firms
leverage. Moreover, the leverage of medium-large firms shows a negligible positive
relationship with capital turnover while small firms show a strong negative relation
between capital turnover and leverage which is consistent with the assumption that the 267
faster the investment turns the lower the need for funding; the sensibility of leverage to
capital turnover is small, as we can see from the coefficient. A model specification that
takes into account the net working capital, improves dramatically the explanatory
power of the model. An increase in net working capital of 10 percent is associated with
a reduction in leverage of 3.6 percent for family firms and of 3 percent for non-family
ones. This demonstrates that investing in working capital components is a key factor,
linked to the choice of capital structure and debt.
These differences between family and non-family medium-large firms, the positive
relationship that exists between management ownership and leverage (Ellul, 2010),
which is higher for family firms and the family firms’ characteristics (tangibility,
depreciation, profitability and inventory) that emerge through the descriptive
statistics, reaffirm that POT alone cannot fully explain family firms’ financial choices
that are affected by governance-related factors arising from the particular interaction
between family and business, and which have an influence on operational management
and motivation to control.
A cross-sectional analysis implies a loss of information on the temporal dynamics of
the variables under consideration. Therefore, we carry out a panel data analysis in
order to verify the cross-sectional results taking into account the time dimension of our
dataset. To test whether a fixed effects or random effects model is appropriate we use a
Hausman test. The results of the specification test in Table V – panel A indicates
that the hypothesis of random effects is rejected at the standard 1 percent level.
To obtain better estimates than traditional ordinary least squares we perform pooling
regressions with adjusted standard errors, i.e. standard errors adjusted to account for
possible correlation within a cluster. In our procedure the clusters are the firms. The
resulting standard errors are unbiased and produce correct confidence intervals
whether the firm effect is temporary or permanent (Petersen, 2009). To deal with the
presence of a industry effect we should control for this parametrically and we choose to
include sector dummies in the model. The results in Table V are substantially the same
we obtained with the cross-sectional analysis with heteroskedasticity robust standard
errors. In summary, the results show that the traditional financial explanatory
variables are relevant in explaining the capital structure choices of both family and
non-family firms and the same result holds for some of the new financial variables we
introduced, namely capital turnover and net working capital. However, family and
non-family firms are also different, the model explanatory power is greater for family
firms as can be seen from the coefficients and statistical significance of market
share (Gallo and Villaseca, 1996), profitability, liquidity and net working capital and
the model R 2. Finally, for both family and non-family medium-large firms, an active
ownership presence in management has a sizeable effect on the capital structure
choices, and this finding is statistically stronger in the case of family members
servicing as CEOs. This result confirms the hypothesis that firms with family
 MF
40,3

268

Table V.
Pooled regressions with

clustered standard errors

Hausman and Chow tests,
Medium large firms
Family Non-family Small firms
Own $ 20 FMngt¼ 1 Own , 20 FMngt¼ 0 Family Non-family

Panel A: pooled results with Hausman test for fixed or random effects
Interc. 0.238 (6.22) 0.256 (5.78) 0.244 (4.91) 0.279 (7.17) 2 0.383 (2 2.91) 20.053 (2 0.20)
LnS 0.001 (0.40) 0.05 (1.28) 2 0.02 (20.44) 2 0.004 (21.26) 0.091 (6.10) 0.046 (1.52)
FMS 2 1.141 (25.35) 2 1.079 (24.13) 2 0.587 (21.88) 2 0.863 (23.88) 2 2.104 (2 0.46) 211.597 (2 2.23)
Prof 2 0.202 (25.48) 2 0.228 (24.91) 2 0.194 (25.48) 2 0.177 (25.69) 2 0.157 (2 3.99) 0.017 (0.48)
Tang 2 0.015 (20.57) 2 0.059 (22.24) 0.009 (0.28) 0.036 (1.24) 2 0.162 (2 7.17) 20.032 (2 0.71)
Liq 2 0.512 (214.10) 2 0.609 (213.33) 2 0.173 (24.30) 2 0.203 (26.04) 2 0.296 (2 9.75) 20.263 (2 6.10)
Inv 0.040 (1.68) 0.012 (0.43) 0.035 (0.86) 0.067 (2.11) 0.146 (7.60) 0.143 (3.41)
CTR 2 0.014 (23.30) 2 0.07 (21.18) 2 0.023 (23.90) 2 0.024 (25.12) 2 0.094 (2 16.41) 20.073 (2 8.26)
FMngt 0.052 (6.38) 0.098 (3.13)
NWC 2 0.203 (211.92) 2 0.222 (210.99) 2 0.163 (27.26) 2 0.173 (29.51) 2 0.367 (2 20.32) 20.310 (2 10.85)
Dindustry Yes Yes Yes Yes Yes Yes
Obs. 18,418 13,560 7,792 12,650 11,162 3,618
R2 0.21 0.24 0.13 0.14 0.38 0.35
Hausman test 21.09 [0.0001] 21.08 [0.0001] 15.06 [0.0001] 16.22 [0.0001] 15.42 [0.0001] 17.78 [0.0001]
Panel B: pooled results with Chow tests for intercept and slope effects (2007 financial crisis)
Interc. 0.187 (4.58) 0.224 (4.66) 0.206 (4.02) 0.231 (5.69) 2 0.675 (2 4.16) 20.236 (2 0.71)
LnS 0.004 (1.12) 0.006 (1.57) 2 0.001 (20.37) 2 0.002 (20.63) 0.120 (6.53) 0.064 (1.72)
(continued)
 Medium large firms
Family Non-family Small firms
Own $ 20 FMngt¼ 1 Own , 20 FMngt¼ 0 Family Non-family

FMS 2 0.996 (24.66) 2 0.898 (23.38) 2 0.505 (21.62) 2 0.792 (23.67) 2 1.420 (2 0.32) 213.467 (2 2.49)
Prof 2 0.160 (23.61) 2 0.206 (23.60) 2 0.168 (24.29) 2 0.143 (24.03) 2 0.195 (2 4.68) 20.025 (2 0.46)
Tang 0.035 (1.16) 2 0.034 (21.06) 0.037 (1.08) 0.082 (2.57) 2 0.044 (2 1.70) 0.030 (0.58)
Liq 2 0.563 (213.82) 2 0.645 (212.95) 2 0.185 (23.77) 2 0.242 (26.04) 2 0.274 (2 8.80) 20.285 (2 6.15)
Inv 0.047 (1.87) 0.007 (0.23) 0.069 (1.68) 0.096 (2.98) 0.175 (6.75) 0.225 (4.15)
CTR 2 0.013 (22.85) 2 0.06 (21.24) 2 0.017 (22.73) 2 0.020 (23.97) 2 0.091 (2 15.69) 20.074 (2 8.48)
FMngt 0.053 (6.25) 0.107 (3.27)
NWC 2 0.184 (29.84) 2 0.211 (29.63) 2 0.153 (26.37) 2 0.153 (27.76) 2 0.323 (2 17.07) 20.276 (2 8.92)
Dindustry Yes Yes Yes Yes Yes Yes
D2007-10 0.158 (3.91) 0.103 (2.24) 0.123 (2.17) 0.175 (3.98) 1.065 (5.45) 0.603 (1.90)
Dslope Yes Yes Yes Yes Yes Yes
Chow test 10.84 [0.0001] 5.09 [0.0001] 6.65 [0.0001] 12.34 [0.0001] 24.93 [0.0001] 3.85 [0.0001]
Obs. 18,418 13,560 7,792 12,650 11,162 3,618
R2 0.21 0.24 0.13 0.14 0.40 0.36
Notes: The medium-large firms sample consist of all non-financial firms in the BVD Aida database, covering the time span 2001-2010, extant in 2010,
with sales . 70 ml e, e, the small firms sample represents firms with sales in the range 5-10 ml e; the table presents parameter estimates from pooling
regressions of book leverage; the t-statistics from clustered standard errors (firms) are in parentheses; family firms are defined in two ways: (1) family-
owned firms in which one or more family members are individually or as a group blockholders, with a minimum threshold ownership of 20 percent; (2)
family controlled and managed firms whose CEO and eventually some other executive director is a member of the blockholding family; book leverage
(BLev) is the ratio of total financial debt to total assets; firm size is measured as the log of sales (LnS); market share (FMS) is the ratio of firm revenues to
industry total revenues; profitability (Prof) is the ratio of operating income before depreciation to total assets; tangibility (Tang) is the ratio of net property,
plant and equipment to total assets; liquidity (Liq) is the ratio of liquid assets to total assets; inventory (Inv) is the ratio of inventory to current assets;
capital turnover (CTR) is the ratio of sales to capital employed; FMngt is a dummy that in family firms takes value one if one or more shareholders in active
management are family members and zero otherwise, in non-family firms it takes value one if a shareholder is in active management and zero otherwise;
net working capital (NWC) is the ratio of current assets minus current liabilities to total assets; to salvage space the coefficients for the industry dummies
and the slope interaction variables are not shown; the Hausman and Chow reported statistics tests, respectively, for fixed vs random effects and for a
structural break triggered by the 2007 financial crisis; t statistics in parentheses, p-values in brackets
Capital structure

Table V.
firms
choices of family

269
MF members involved in active management are more levered than family firms
40,3 professionally managed. This finding is in contrast with the agency theory for which the
presence of the family in the board would bring a substitution effect, i.e. more direct
monitoring of management by family board members implies a reduced need to use debt
to prevent managerial opportunistic behavior (González et al., 2013). Our result is
consistent with the socioemotional framework, which contradicts some basic agency
270 theory prediction (Berrone et al., 2012). Gomez-Mejia et al. (2007) suggests a relation
between the role of the family in the business and the desire to protect its socioemotional
wealth by means of the preservation of control. The presence of the family in active
management signals the family commitment to the business and makes stronger the
sense of self-identification with the firm which is a relevant dimension of the
socioemotional wealth framework (Gomez-Mejia et al., 2007; Berrone et al., 2012).
We observe that the balance sheet variables present a higher explanatory power for
small family firms; the model specification, in this case, does not include the presence
of the family in active management because this data is unavailable in AIDA, but
nonetheless it explains 38 percent of the time series and cross-sectional variability.
Moreover, small family firms’ leverage has a strong relation with size, but it is not
affected by the firm’ market share.
Given that our dataset covers the period 2001-2010 an important issue is to analyse
whether the recent financial crisis influenced substantially the leverage decisions of
family and non-family firms. To verify this hypothesis we introduce in the model an
intercept dummy that takes value one for the period 2007-2010 and interaction variables
to account for differential relationships with the dependent variables in the same period.
The results for the Chow F tests show that we can reject the null hypothesis that the
intercept and slope dummy variables are jointly zero. The conclusion is that the
estimated model has different parameter values before and after the financial crisis,
another interesting observation we can draw form the results is that the effect of the
crisis is stronger for family than non-family firms, but the medium-large family firms
were in better conditions to cope with this structural change with respect to the small
family firms.

7. Conclusions
The present study focuses on the determinants of capital structure choices of
non-financial medium-large private family firms and how family governance-related
factors impact on them. Family firms present a higher leverage level than non-family
firms, their ownership is strictly domestic and, consistent with the main theories
explaining family firms’ behavior, and they are more profitable although their firm
market share is lower. Family firms are smaller in size and characterized by a higher
level of tangibility and higher turnover of the capital employed. These overall
characteristics tend to persist at the industry level. We show that capital structure
choices of medium-large family firms are linked to governance characteristics and to
variables related to the balance sheet structure that have not been used in previous
studies, such as net working capital and capital turnover. Our findings highlight that
family firms avoid opening their capital to non-family shareholders, as “family control
and influence” is an important dimension of the family socioemotional wealth. So
family firms mitigate risk aversion and finance their growth using debt but at the same
time family firms with leading market share positions in their industry recur less to
debt financing maybe as a result of their increased capability to produce cash flow. Capital structure
Leverage tends to increase with family ownership and family involvement in choices of family
management. Family firms are more levered than the non-family counterpart and the
differences are economically and statistically more significant for medium-large than firms
small firms. This suggests the small size of family firms could not be the result of debt
aversion but the consequence of constraints on credit availability. This is supported by
the evidence of an impact of the recent financial crisis on family firms, stronger for 271
small businesses.
Our results, based on a large dataset, can give public authorities an important
insight in order to facilitate firms funding specially in the current critical economic
scenario, relaxing the regulation to permit also to private firms to issue bonds or
promoting the development of credit funds as a viable alternative to bank financing.
An international comparison could be the next step in developing this research, in
order to check if the results are influenced by country-related regulatory aspects, and
national cultural aspects, which could affect the control dimension of the
socioemotional wealth. Moreover, further studies should be developed on the impact
of the other dimensions – in the first the renewal of family bonds to the firm – on
capital structure choices of private firms. An interesting area of research could be the
integration of the socioemotional wealth dimensions with the law and finance approach
to take into account the level of protection of investors and the rights of shareholders
and creditors, to develop a comprehensive model of family firms’ financial decisions.
The main difficulty for this research agenda remains the availability of data and
information on private firms.

Note
1. The results for this model specification are available on request.

References
Ampenberger, M., Bennedsen, M. and Zhu, H. (2012), “The capital structure of family firms”,
in Cumming, D. (Ed.), Oxford Handbook of Entrepreneurial Finance, Oxford University
Press, Oxford, pp. 167-191.
Ampenberger, M., Schmid, T., Achleitner, A.K. and Kaserer, C. (2013), “Capital structure
decisions in family firms: empirical evidence from a bank-based economy”, Review of
Managerial Sciences, Vol. 7 No. 3, pp. 247-275.
Anderson, R.C. and Reeb, D.M. (2003a), “Founding-family ownership and firm performance:
evidence from the S&P 500”, Journal of Finance, Vol. 58 No. 3, pp. 1301-1328.
Anderson, R.C. and Reeb, D.M. (2003b), “Founding-family ownership, corporate diversification
and firm leverage”, Journal of Law & Economics, Vol. 46 No. 2, pp. 653-680.
Anderson, R.C., Duru, A. and Reeb, D.M. (2010), “Family preferences and investment policy:
evidence from R&D spending and capital expenditures”, working paper, Kogod School of
Business, American University, Washington, DC.
Anderson, R.C., Mansi, A.M. and Reeb, D.M. (2003), “Founding family ownership and the agency
cost of debt”, Journal of Financial Economics, Vol. 68 No. 2, pp. 263-285.
Andres, C. (2008), “Large shareholders and firm performance: an empirical examination of
founding-family ownership”, Journal of Corporate Finance, Vol. 14 No. 4, pp. 431-445.
MF Antoniou, A., Guney, Y. and Paudyal, K. (2008), “The determinants of capital structure: capital
market-oriented versus bank-oriented institutions”, Journal of Financial & Quantitative
40,3 Analysis, Vol. 43 No. 1, pp. 59-92.
Barontini, R. and Caprio, L. (2006), “The effect of family control on firm value and performance:
evidence from continental Europe”, European Financial Management, Vol. 12 No. 5,
pp. 689-723.
272 Bauguess, S. and Stegemoller, M. (2008), “Protective governance choices and the value of
acquisition activity”, Journal of Corporate Finance, Vol. 14, pp. 550-566.
Berrone, P., Cruz, C. and Gomez-Mejı́a, L.R. (2012), “Socioemotional wealth in family firms:
theoretical dimensions, assessment approaches, and agenda for future research”, Family
Business Review, Vol. 25, pp. 258-279.
Blanco-Mazagatos, F., de Quevedo-Puente, E. and Castrillo, L.A. (2007), “The trade off between
financial resources and agency costs in the family business: an exploratory study”, Family
Business Review, Vol. 20 No. 3, pp. 199-213.
Brav, O. (2009), “Access to capital, capital structure, and the funding of the firm”, Journal of
Finance, Vol. 64 No. 1, pp. 263-308.
Caprio, L. and Croci, E. (2011), “Ownership structure, family control, and acquisition decisions”,
Journal of Corporate Finance, Vol. 17 No. 5, pp. 1636-1657.
Carney, M. (2005), “Corporate governance and competitive advantage in family-controlled firms”,
Entrepreneurship Theory & Practice, Vol. 29, pp. 249-265.
Chami, R. (1999), “What’s different about family business?”, working paper, University of Notre
Dame and the International Monetary Fund, Washington, DC.
Cohen, G. and Yagil, J. (2010), “Sectorial differences in corporate financial behavior:
an international survey”, European Journal of Finance, Vol. 16 No. 3, pp. 245-262.
Coleman, S. and Carsky, M. (1999), “Sources of capital for small family-owned businesses
evidence from the national survey of small business finances”, Family Business Review,
Vol. 12 No. 1, pp. 73-84.
Correia da Silva, L., Goergen, M. and Renneboog, L.D.R. (2004), Dividend Policy and Corporate
Governance, Oxford University Press, Oxford.
Croci, E., Doukas, J. and Gonenc, H. (2011), “Family control and financing decisions”, European
Financial Management, Vol. 17 No. 5, pp. 860-897.
Dahya, J., Dimitrov, O. and McConnell, J. (2008), “Dominant shareholders, corporate boards,
and corporate value: a cross-country analysis”, Journal of Financial Economics, Vol. 87,
pp. 73-100.
Daily, C.M. and Dollinger, M.J. (1992), “An empirical examination of ownership structure in
family and professionally managed firms”, Family Business Review, Vol. 5 No. 2,
pp. 117-136.
Donaldson, G. (1961), Corporate Debt Capacity: A Study of Corporate Debt Policy and the
Determination of Corporate Debt Capacity, Division of Research, Harvard Graduate School
of Business Administration, Boston, MA.
Ellul, A. (2010), “Control motivations and capital structure decision”, working paper, Kelley
School of Business, Indiana University, Bloomington, IN.
Faccio, M. and Lang, L.H.P. (2002), “The ultimate ownership of Western Europe corporations”,
Journal of Financial Economic, Vol. 65 No. 3, pp. 365-395.
Fama, E.F. and Jensen, M.C. (1983), “Agency problems and residual claims”, Journal of Law
& Economics, Vol. 26, pp. 325-344.
Frank, M.Z. and Goyal, W.K. (2009), “Capital structure decisions: which factors are reliably Capital structure
important?”, Financial Management, Vol. 38 No. 1, pp. 1-37.
choices of family
Franks, J., Mayer, C., Volpin, P. and Wagner, H.F. (2010), “The life cycle of family ownership;
a comparative study of France, Germany, Italy and UK”, London Business School, London. firms
Fulghieri, P. and Suominen, M. (2012), “Corporate governance, finance, and the real sector”,
Journal of Financial and Quantitative Analysis, Vol. 47 No. 6, pp. 1187-1214.
Gallo, M.A. and Vilaseca, A. (1996), “Finance in family business”, Family Business Review, Vol. 9 273
No. 4, pp. 387-401.
Gomez-Mejia, L.R., Makri, M. and Kintana, M.L. (2010), “Diversification decisions in
family-controlled firms”, Journal of Management Studies, Vol. 47 No. 2, pp. 223-252.
Gomez-Mejia, L.R., Welbourne, T.M. and Wiseman, R.M. (2000), “The role of risk sharing and
risk taking under gainsharing”, Academy of Management Review, Vol. 25 No. 3,
pp. 492-507.
Gomez-Mejia, L.R., Haynes, K.T., Nunez-Nickel, M., Jacobson, K.J.L. and Moyano-Fuentes, J.
(2007), “Socioemotional wealth and business risks in family-controlled firms: evidence
from Spanish olive oil mills”, Administrative Science Quarterly, Vol. 52 No. 1, pp. 106-137.
González, M., Guzmán, A., Pombo, C. and Trujillo, M.A. (2013), “Family firms and debt: risk aversion
versus risk of losing control”, Journal of Business Research, Vol. 66 No. 11, pp. 2308-2320.
Graham, J.R. and Leary, M.T. (2011), “A review of empirical capital structure research and
directions for the future”, Annual Review of Financial Economics, Vol. 3, pp. 309-345.
Gugler, K. (2003), “Corporate governance, dividend payout policy, and the interrelation between
dividends, R&D, and capital investment”, Journal of Banking & Finance, Vol. 27 No. 7,
pp. 1297-1321.
Hagelin, N., Holmén, M. and Pramborg, B. (2006), “Family ownership, dual-class shares, and risk
management”, Global Finance Journal, Vol. 16, pp. 283-301.
Harris, D., Martinez, J.I. and Ward, J.L. (1994), “Is strategy different for the family-owned
business?”, Family Business Review, Vol. 7, pp. 159-174.
Haynes, G., Walker, R., Rowe, B. and Hong, G. (1999), “The intermingling of business and family
finances in family owned businesses”, Family Business Review, Vol. 12, pp. 225-239.
Hölmstrom, B. (1979), “Moral hazard and observability”, Bell Journal of Economics, Vol. 10 No. 1,
pp. 74-91.
Jensen, M.C. and Meckling, W.H. (1976), “Theory of the firm: managerial behavior, agency costs
and ownership structure”, Journal of Financial Economics, Vol. 3 No. 4, pp. 305-360.
King, M.R. and Santor, E.B. (2008), “Family values: ownership structure and performance of
Canadian firms”, Journal of Banking & Finance, Vol. 32 No. 11, pp. 2423-2432.
Klasa, S. (2007), “Why do controlling families of public firms sell their remaining ownership
stake?”, Journal of Financial and Quantitative Analysis, Vol. 42 No. 2, pp. 339-367.
La Porta, R., Lopez-de-Silanes, F. and Shleifer, A. (1999), “Corporate ownership around the
world”, Journal of Finance, Vol. 54 No. 2, pp. 471-517.
Lemmon, M.L. and Zender, J.F. (2010), “Debt capacity and tests of capital structure theories”,
Journal of Financial and Quantitative Analysis, Vol. 45 No. 5, pp. 1161-1187.
Lòpez-Gracia, J. and Sànchez-Andujar, S. (2007), “Financial structure of the family businesses:
evidence from a group of small Spanish firm”, Family Business Review, Vol. 20, pp. 269-287.
McConaughy, D.L. and Philipps, G.M. (1999), “Founders versus descendants: the profitability,
efficiency, growth characteristics and financing in large, public, founding
family-controlled firms”, Family Business Review, Vol. 12, pp. 123-132.
MF Marchica, M.T. and Mura, R. (2010), “Financial flexibility, investment ability, and firm value:
evidence from firms with spare debt capacity”, Financial Management, Vol. 39 No. 4,
40,3 pp. 1339-1365.
Margaritis, D. and Psillaki, M. (2010), “Capital structure, equity ownership and firm
performance”, Journal of Banking & Finance, Vol. 34, pp. 621-632.
Miller, D., Le Breton-Miller, I. and Lester, R.H. (2011), “Family and lone founder ownership and
274 strategic behaviour: social context, identity, and institutional logics”, Journal of
Management Studies, Vol. 48 No. 1, pp. 1-25.
Mishra, C.S. and McConaughy, D.L. (1999), “Founding family control and capital structure: the
risk of loss of control and the aversion to debt”, Entrepreneurship Theory & Practice,
Vol. 23 No. 4, pp. 53-64.
Molly, V., Laveren, E. and Deloof, M. (2010), “Family business succession and its impact on
financial structure and performance”, Family Business Review, Vol. 23, pp. 131-147.
Molly, V., Laveren, E. and Jorissen, A. (2012), “Intergenerational differences in family firms:
impact on capital structure and growth behaviour”, Entrepreneurship Theory & Practice,
Vol. 36 No. 4, pp. 703-725.
Myers, S.C. and Majluf, N.S. (1984), “Corporate financing and investment decisions when firms
have information that investors do not have”, Journal of Financial Economics, Vol. 13 No. 2,
pp. 187-221.
Petersen, M.A. (2009), “Estimating standard errors in finance panel data sets: comparing
approaches”, Review of Financial Studies, Vol. 22 No. 1, pp. 435-480.
Rajan, R.G. and Zingales, L. (1995), “What do we know about capital structure? Some evidence
from international data”, Journal of Finance, Vol. 50 No. 5, pp. 1421-1460.
Romano, C.A., Tanewski, G.A. and Smyrnios, K.X. (2001), “Capital structure decision making:
a model for family business”, Journal of Business Venturing, Vol. 16 No. 3, pp. 285-310.
Sacristán-Navarro, M., Gómez-Ansón, S. and Cabeza-Garcı́a, L. (2011), “Family ownership and
control, the presence of other large shareholders, and firm performance: further evidence”,
Family Business Review, Vol. 24 No. 1, pp. 71-93.
Schulze, W.S., Lubatkin, M.H. and Dino, R.N. (2002), “Altruism, agency, and the competitiveness
of family firms”, Managerial and Decision Economics, Vol. 23 Nos 4/5, pp. 247-259.
Schulze, W.S., Lubatkin, M.H., Dino, R.N. and Buchholtz, A. (2001), “Agency relationships in
family firms. Theory and evidence”, Organization Science, Vol. 12, pp. 99-116.
Setia-Atmaja, L., Tanewski, G. and Skully, M. (2009), “The role of dividends, debt and board
structure in the governance of family controlled firms”, Journal of Business Finance
& Accounting, Vol. 36 Nos 7/8, pp. 863-898.
Sheikh, N.A. and Wang, Z. (2011), “Determinants of capital structure: an empirical study in firm
in manufacturing industry of Pakistan”, Managerial Finance, Vol. 37 No. 2, pp. 117-133.
Shyam-Sunder, L. and Myers, S.C. (1999), “Testing static trade off against pecking order models
of capital structure”, Journal of Financial Economics, Vol. 51, pp. 219-244.
Sraer, D. and Thesmar, D. (2007), “Performance and behavior of family firms: evidence from
the French stock market”, Journal of the European Economic Association, Vol. 5 No. 4,
pp. 709-751.
Tagiuri, R. and Davis, J.A. (1992), “On the goals of successful family companies”, Family
Business Review, Vol. 5, pp. 43-62.
Titman, S. (1984), “The effect of capital structure on a firm’s liquidation decision”, Journal of
Financial Economics, Vol. 13, pp. 137-151.
Villalonga, B. and Amit, R. (2006), “How do family ownership, control and management affect Capital structure
firm value?”, Journal of Financial Economics, Vol. 80 No. 2, pp. 385-417.
Wiseman, R.M. and Gomez-Mejia, L.R. (1998), “A behavioral agency model of managerial risk
choices of family
taking”, Academy of Management Review, Vol. 23 No. 1, pp. 133-153. firms
About the authors
Pietro Gottardo graduated in economics at the University of Pavia (Italy). He achieved the 275
Doctorate in “Financial Markets” at the University of Bergamo (Italy). He is an Associate Professor
in “Corporate Finance”. Main research interests are capital structure, trading and financial
markets, risk management. current teaching: “Corporate Finance”, “Financial Modeling”, at the
Department of Economics and Management of the University of Pavia.
Anna Maria Moisello graduated in economics at the University of Pavia (Italy). She achieved
the Doctorate in “Business and Administration” at the “Bocconi University” of Milan (Italy). She is
a Researcher in “Business and Administration”; main research interests in management accounting
and control. Current teaching: “Business and Administration”, “Management Control” at the
Department of Economics and Management of the University of Pavia. Anna Maria Moisello is the
corresponding author and can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.