Capital Structure Policy and Determinants:

Evidence from the Portuguese Banking Industry

Manuel O. Marques

Associate Professor

Faculdade de Economia do Porto

Rua Roberto Frias 4200 Porto Portugal

E-mail: momarques@fep.up.pt

Mário C. Santos

Visiting Assistant Professor

DEGEI - Universidade de Aveiro

Campo de Santiago - 4150 Aveiro Portugal

E-mail: msantos@egi.ua.pt

(Preliminary draft – this version November 2, 2003)

Comments are welcome

Please do not quote or cite without permission

- 2 -

Capital Structure Policy and Determinants:

Evidence from the Portuguese Banking Industry

ABSTRACT

The paper examines theoretically and explores empirically the problem of the banking

firm’s capital structure (voluntary) decisions. Data was gathered through a survey conducted to a

sample of 89.5 percent of the Chief Executive Officers (CEOs) of Portuguese banks in office

during the 1989-1998 period.

Findings support the notion that Portuguese banks’ debt/equity choice do matter. They

also indicate that surveyed CEOs show a relative preference for the trade-off capital structure

policy model. Survey results are consistent with a number of theoretical propositions typically

associated with the determinants of debt-equity choice of non-financial firms. Specifically,

evidence supports that factors associated with the role of debt and non-debt tax shields at the

bank level, as well as agency and governance conflicts and asymmetric information

considerations are relevant for capital structure managerial decision-making of Portuguese

banks.

Overall, findings provide support to the notion that the design and the adjustment of the

firm’s financial structure may be explained within the framework of both theoretical and

empirically motivated determinants well established in capital structure literature.

Keywords: Capital structure policy, Portuguese banking industry, Survey,

JEL Classification: C42; G21; G32

- 3 -

1. INTRODUCTION1

The problem of how firms choose and adjust their strategic financing mix has called a

great deal of attention from corporate financial economists, and has been a source of intense

debate. At the outset of such debate is, among others issues, the question of the relevance of firm

strategic financing decisions for its valuation. Not surprisingly, large and self-contained bodies

of both theoretical and empirical research developed in the literature. Among these strands of

literature, the firm’s capital structure emerged as one of the most researched topics in corporate

finance.2

Despite some significant contributions to the general perception of the various intricacies

about corporate capital structure, research produced so far did not provide yet a sound basis for

establishing, in a decisive fashion, the empirical validity of the different theoretical models.

Probably the most eclectic, prevalent and non-controversial view, with respect to the contention

surrounding the corporate capital structure theory, is Myers’s argument that it is a puzzle,3

mirrored by Kamath’ (1997) enigma, Stiglitz’ dilemma4, or, as suggested in The Economist, a

mystery.5

It appears that (1) we are still lacking a comprehensive theory to explain how firms

decide about their strategic financing; and (2) yet we can not unambiguously specify the relation

between capital structure choice and firm value.

Since the foundational work of Modigliani and Miller (1958), a number of authors extended their

capital structure irrelevancy theory. The literature also thoroughly describes the various attempts

to model corporate debt/equity policy. However, what optimal mix of securities should a firm

issue still remains undetermined.

If extant capital structure theoretical literature has so far successfully modeled a «large

number of potential determinants of capital structure» choice (Harris and Raviv 1991), empirical literature

has as well failed in finding unambiguous and compelling validation of the contextual relevance

of such models. Available empirical evidence often appears to show significant dependence of

the observed reality and the research methods applied, leading sometimes to unconvincing and 1 The paper draws partially on the second author’s PhD dissertation. We are deeply grateful to Aníbal Santos, Pedro Duarte Silva, Ricardo Cruz, and Sam Hayes for valuable suggestions and comments. The user disclaimer applies concerning errors and omissions. 2 Other important research areas are, for example, security design, and debt maturity structure choice. 3 See Myers (1984). 4 See Stiglitz (1989).

- 4 -

contradictory results.6 As suggested by Frankfurter and Philippatos (1992) one of the debilities of

corporate finance theories «is their weak correspondence to facts».7

Overall we still lack a satisfactory, comprehensive and positive explanation for firms’

capital structure observed behavior. Theoretically, it is still not well understood why firms’

financial contracts recurrently appear in certain patterns (e.g., Harris and Raviv 1989). This

suggests that we need to resort to a more robust framework to gather useful insights into the

financing behavior of actual real-world firms.

Capital structure literature recurrently highlights the topic at the non-financial firm level.8

Further, corporate finance empirical research has, to a certain extent, frequently disregarded

financial industry data (e.g., Fama and French 1992). Overall it seems that the investigation of

capital structure of financial firms such as banks has been largely overlooked.9

As suggested by Chen and Mazumdar (1994) «the bank capital structure debate remains

unresolved.» However, as posited, e.g, by Orgler and Taggart (1983) general capital structure

theory could provide «a useful framework for analyzing bank capital structure.» Merton Miller’s (1995)

question if «M&M propositions apply to banks?» is also consistent with the view that banks’ debt/equity

choice still remains an empirical question. On these grounds, the banking firm capital structure

problem appears as a promising topic for empirical research to enhance our understanding of the

capital structure puzzle, through the investigation of an extreme financial leverage polar-case

(Masulis 1988).10

It is a truism that the mainstream of theoretical and empirical research on corporate

capital structure springs from the examination of U.S. phenomena. This fact hinders the

generalization of these results to other countries or geographical areas with (sometimes)

5 See, The Economist, January 6th 1996, p. 61. 6 An interesting illustration is provided by Opler and Titman (1996) who suggest that in Fischer, Heinkel and Zechner’s (1989) capital structure (dynamic) model, «a firm issues equity after its share price declines and repurchases equity after its share prices increase to adjust towards an optimal capital structure.» Extant empirical and anecdotal evidence indicates that firms actually do the opposite. 7 One plausible explanation for this phenomenon may be the misalignment between the behavioral characteristics of financing choices available to firms and the theoretical microeconomic underpinnings of the standard neoclassical model of the firm, which represents a general theoretical foundation for a number of corporate finance models. 8 As observed by Dowd (1996) «traditional banking literature tends to overemphasize the difference between banks and non-financial firms, and therefore overlooks important similarities between them.» 9 To the best of our knowledge Marcus (1983), Sharpe (1995), Osterberg and Thomson (1996), and Hasan (1997) are some of the very few papers on this topic. 10 According to Masulis (1988) “one means of expanding on the previous evidence [on capital structure] is to study the determinants of leverage in an industry exhibiting extreme leverage choices. Some of the best examples of this situation are commercial banks […] which typically have leverage ratios of 95 percent debt to assets and higher.”

- 5 -

remarkably dissimilar economic, financial, and institutional conditions what, in these conditions,

would seem inappropriate or even imprudent.11 Among these conditions are, as noted by

Milgrom and Roberts (1992), «… rights that come with ownership vary among countries and over time.»

Allocative functions of financial markets may also vary widely across countries. Thus,

informational and operating efficiency, and liquidity, are institutional features of financial

markets that may have a role as “determinant(s) of corporate financing choices” (Demirgüç-Kunt and

Maksimovic 1996).12

According to Saá-Requejo (1996), Rajan and Zingales (1995), and Harris and Raviv

(1992), among others, further substantiation of capital structure hypotheses is needed to increase

the robustness of their predictions. This desideratum may be pursued through their empirical

testing in different environmental contexts of country, time and industry. Such investigations

may be helpful for a better understanding of the implications of environmental and behavioral

factors on capital structure decisions, and thus contributing for broadening the explanatory and

predictive power of the theory.

Field-based research methods are a long-established practice in corporate finance

investigation.13 However it seems to have recently experienced a renewed interest on its use with

the argument that bringing together both "traditional" and field-based types of research for

«studying the activities of a few firms may be one of the only ways to study phenomena which are not easily

quantifiable.»14, 15

In order to evaluate how standard capital structure theory could handle strategic financing

decisions of Portuguese banks we conducted a survey to a sample of CEOs through a person-to-

person interview supported by a structured questionnaire.

11 Those conditions include: market structures; regulatory frameworks; governance systems including relations between banks and firms; cultural environment; accounting principles and practices; legal system including tax and bankruptcy laws; financial system design, degree of development, and regulation; national savings levels; and risk preferences of capital market participants. 12 Remarkable disparities among countries relate, among other dimensions, to firms’ characteristics (e.g., size), market structures, legal and regulatory regimes, prevailing governance systems, cultural environments, accounting principles and practices, which might well be relevant sources of variance across space, time and industry. 13 Lintner (1956), and Donaldson (1961) provide good illustrations. 14 Foddy (1993) suggests that «asking questions is widely accepted as a cost-efficient (and sometimes the only) way, of gathering information about past behaviour and experiences, private actions and motives and attitudes (i.e. subjective variables that cannot be measured directly).» See also Jensen et al. (1989). 15 The Board of Governors of the Federal Reserve System has conducted “The National Survey of Small Business Finances” in 1987, 1993 and 1998. Survey data sets have been made available to academia and extensively used, among others, in Ang, Cole and Lin (2000), Berger and Udell (1998) and Petersen and Rajan (1994).

- 6 -

Our empirical approach to the studying of the bank capital structure problem, although

having a similar research objective of other recent survey-based studies, e.g., Graham and

Harvey (2001), and Bancel and Mittou (2002), distinguishes from this prior research in a number

of aspects.

First, we conducted the survey in a face-to-face interview format rather than administered

by mail, as is the case with the majority of prior studies.16 Our empirical research methodology

minimizes some of the potential problems that the survey method may suffer, namely, non-

response and response biases. We also reduced the potential for survey participants interpreting

survey questions differently, since interviews were personally conducted by one of the authors.

Secondly, the study was conducted within a single and relatively homogeneous industry,

thus avoiding difficulties in controlling the unsystematic effects, inevitably present in cross-

sectional samples.

Third, we surveyed approximately 90 percent of the population, comparing to an average

response rate of 20.3 percent in capital structure mail-administered surveys.17

Fourth, although confidentiality and anonymity were guaranteed to survey participants,

we were able to control across important bank characteristics: ownership, position in the life

cycle, listing condition and capitalization.

Fifth, the survey was conducted outside the U.S. and consequently offers an opportunity

for generalizing empirical results obtained for the propositions submitted to test.

Lastly, in our study we combine qualitative survey data and quantitative data drawn from

a database for banks financials developed by the authors.

In this investigation we aim at providing evidence to the following generic research questions:

(1) Are capital structure theories incorporated in CEOs’ decisions? (2) Would capital structure

managerial decision-making provide empirical support for extant theories? (3) Which are the

potential determinants of debt / equity managerial policies?

The remaining of the paper is organized as follows: Section two introduces and discusses

the theoretical background of the capital structure problem at the firm level, building the basis

for the development of the testable propositions included in the survey instrument. Section three

examines methodological issues related to our empirical study, and describes survey design, 16 See Appendix 2.1 to Chapter 2 and Appendix 5.4 to Chapter 5 for summaries of survey-based research in corporate capital structure.

- 7 -

sample selection and data. Section four reports survey results. Section five summarizes and

concludes the paper.

2. THEORETICAL BACKGROUND

The undertaking of risky investment projects increasingly requires larger pooling of

financing. Such amounts of resources are frequently beyond a firm’s ability to generate and

retain cash. To cope with this potential shortage of financial capital, firms increasingly tend to

organize as larger and more complex business organizations.18 A wealth-constrained firm owner

endowed with a profitable investment opportunity is driven to raise external capital to finance the

project by selling securities. These securities vary in terms of claims to issuer’s future rents and

in terms of allocation of residual rights of control.19 The capital structure problem emerges from

the definition of the mix of securities the firm should optimally issue.20

Modigliani and Miller (1958)21 provided the foundational impulse to the study of the

capital structure problem by formally proving that, under conditions of complete, perfect and

frictionless markets, a firm’s market value and the welfare of its security holders remain

unaffected by financing decisions.22 This theoretical proposition carries the implications that: (1)

financing and investment policies are independent; (2) internal and external financing are perfect

substitutes; and (3) the specific type of the financing contractual arrangement, either equity or

debt, is also irrelevant.

M&M’s irrelevance theory was subsequently extended and its results generalized under a

less stringent set of assumptions, showing that the theorem still obtains in the presence of risky 17 See Table 5.4 appended to chapter 5. 18 Since the allocation of funds implied by the increased scale of projects is, typically, out of reach for more primitive forms of business organization, the modern corporation emerges as a much more open vehicle for pooling capital, and with less restricted residual claims. Hansmann (1996) observes that the «large-scale enterprise will be organized in the form of investor-owned firms.» This idea is corroborated by Easterbrook and Fischel (1991) who claims that «publicly held corporations dominate other organizational forms when the technology of production requires firms to combine both the specialized skills of multiple agents and large amounts of capital.» 19 Grossman and Hart (1986) suggest that “contractual rights can be of two types: specific rights and residual rights. When it is costly to list all specific rights over assets in the contract, it may be optimal to let one party purchase all residual rights. Ownership is the purchase of these residual rights.” According to Jensen and Meckling (1976) “there is another important dimension to this problem [capital structure] — namely the relative amount of ownership claims held by insiders (management) and outsiders (investors with no direct role in the management of the firm).” 20 Without loss of generality, capital structure theoretical literature emphatically assumes that investors provide external financing under two major types of contracts: debt and equity. Therefore, a number of characteristics often found in real world corporate security issuance are not considered in this context. 21 M&M hereafter.

- 8 -

debt and hybrid securities, and when the risk class assumption is relaxed. Similarly, the

irrelevance proposition still holds in both single period and intertemporal modeling settings,

implying that the market value of the firm is also unrelated to debt maturity structure.23

Despite its unquestionable analytical elegance, rigor and scientific importance, M&M’s

irrelevancy theory is not a useful tool to either directly explain or predict capital structure

behavior of actual real-world firms.24 However, as claimed by Miller (1992), «showing what doesn’t

matter can also show, by implication, what does».25 Therefore, the «'nothing matters” in corporate finance»

implication of M&M’s theorem should be rationalized as a result obtained under a corporate

setting unaffected by any kind of environmental frictions and market imperfections.26

The prototypical neo-classic firm underlying M&M theory suffers from some congenital

shortcomings, which severely curtail its ability to be a satisfactory tool in the examination of the

financial structure problem of the firm. The neo-classical firm is portrayed as a single economic

agent (or a unanimous group) whose actions follow a specific and pre-determined decision-

making criterion. This implies the presumption that ownership and decision-making are

congruently tied, and there is consequently no role for specialization in ownership and

management functions.27

Furthermore it is assumed, either implicitly or explicitly, that agents have free, complete

and perfect information for their resource allocation choices. Thus, a major weakness of the neo-

classical paradigm is the failure to recognize that specialization in ownership and management

may emerge, and a firm’s owners may accordingly play no effective role in controlling the use of

firm assets at all. This carries, among others, the implication that the neoclassical theory tends to

neglect the productive role of the firm’s management team.

To specify an economic environment able to usefully accommodate some conspicuous

features of real-world firms’ financing behavior, we should though consider withdrawing from 22 Gertler (1988) and Fama (1990), among others, posit that the Arrow and Debreu economic environment underlies the M&M (1958) theorem. 23 See, e.g., Stiglitz (1969, 1974), Fama and Miller (1972), Smith (1972), Baron (1974, 1976), Merton (1974), Hagen (1976), and Hellwig (1981). 24 As pointed out by Ross (1988) the M&M analysis «was designed less to verify some mathematical truism than to capture a live and elusive scientific intuition.» 25 Stewart Myers (1993) corroborates Miller’s view, arguing that «it seems that financial leverage matters more than ever.» See also Stiglitz (1989) who argues that «as long as resources are spent on thinking about (and implementing) a corporate financial strategy, we cannot simultaneously hold to the view that firms are rational profit maximizing (shareholders are rational investors) and that debt-equity ratios (and financial policies more generally) are irrelevant.» 26 As suggested by Myers (1993) the M&M practical message is that «if there is an optimal capital structure, it should reflect taxes or some specifically identified market imperfections.»

- 9 -

two major assumptions of the neo-classical paradigm. The first is the notion that specialization in

ownership and managerial functions is possible under certain contracting circumstances.28

Therefore, the power committed to professional managers over the firm’s decision-making

process, and the controlling power assigned to residual claimants, may in certain instances be

separated.29 In an environment where ownership is separated from control, firm’s residual

claimants who bear the major part of the risk exercise their ownership control rights by

monitoring their agents’ decision-making. Since these do not bear a substantial share of the

potential adverse wealth effects of their decisions, owners’ wealth may be at risk in adverse

states of the world.30 Whenever owners do not control the firm’s management functions, they

cannot directly observe the firm’s true economic characteristics. Thus, owners may be at a

disadvantage, because firm insiders have a potential incentive to opportunistically exploit their

informational advantage.31

The second major departure from the neo-classic firm paradigm relates to the concept

that a firm is a nexus-of-contracts; 32 i.e. complex, an intricate and massive web of contractual

arrangements among claimholders, that co-ordinates productive and other economic

relationships, making the firm a kind of a legal fiction.33

Ownership rights are a central component of this network of contractual relationships.34,

35 The specification of such individual ownership rights requires contracting the allocation of

27 According to Jensen (1983) “there are no ‘people’ problems or information problems” in the neoclassical paradigmatic firm. 28 The acknowledgement that their owners may not operate certain firms is a stylized fact firstly recognized by Adam Smith, and later hypothesized in Berle and Means (1932). Neoclassical economics «does not distinguish managers from owners» (Auerbach 1992). 29 In Berle and Means (1932) control is used as a meaning of performing the functions and exercise authority over the management of a firm. According to Bolton and Scharfstein (1998) «the diffuse equity ownership structure implied in Berle and Means’ argument gives managers effective control of the firm, raising concerns about the possibility of managers to run the firm to their own benefit, possibly at the expense of investors.» 30 In this framework, owners might be willing to trade-off increased risk for the economic rents generated by the firm (Fama and Jensen 1983). They may also engage in hedging strategies. However, as individual portfolios become increasingly less risky, individual owners may experience a desincentive to exercise their (costly) ownership control rights (e.g., Fama 1980). 31 Eggertsson (1990) defines opportunistic behavior as a «breach of contract involving strategic manipulation of information.» 32 Pioneered by Coase (1937) the nexus-of-contracts view of the firm has received widespread support (e.g., Milgrom and Roberts 1992; Jensen and Smith 1985; Jensen and Meckling 1976; and Alchian and Demsetz 1972). Bolton and Scharfstein (1998) and Allen and Winston (1995) note that the nexus-of-contracts is a dominant paradigm in modern corporate finance. 33 Jensen and Smith (1985) argue that «the behavior of an organization is the equilibrium behavior of a complex contractual system made up of maximizing agents with diverse and conflicting objectives.» 34 Stiglitz (1989) argues that control rights are valuable. He illustrates the argument noting that «in some instances there are two kinds of shares that are identical in their claims on the profits of the firm, but one of which is nonvoting (or has fewer voting rights). These shares often sell at a large differential.»

- 10 -

both residual rights of control36 and residual rents (e.g., Milgrom and Roberts 1992; Hansmann

1988). As suggested by Jensen and Meckling (1976), the «individual behavior in organizations […] will

depend upon the nature of these contracts.»

Arguably, the importance of residual control rights derives from the difficulty of writing

complete contracts.37, 38 However, as pointed out by Williamson (1990), writing complete

contracts did not prove to be an effective governing mechanism for economic relationships

between parties whose transactions are contingent on some future state of the nature.39 This

uncertainty, which is an innate characteristic of incomplete contracts, may become a source for

potential opportunistic behavior of parties (Bolton and Scharfstein 1998). Among other types,

opportunistic behavior may assume the form of ex post opportunism in situations characterized

by asset specificity, and the form of hold-up problems whenever there are relationship-specific

investments (Bolton and Scharfstein 1998; Klein, Crawford and Alchian 1978).40

According to the property rights view, capital structure is not a matter of indifference

because financial contracts confer residual rights of control in addition to rights to share the

economic rents.41, 42 Differently, the incomplete contract theory emphasizes the state-contingent

35 According to Hart and Moore (1990) and Grossman and Hart (1986) the nature of firm’s ownership is related to owners’ (residual) control rights over the use of firm’s non-human assets. In this perspective, ownership is akin to control. Contrastingly, Berle and Means (1932) implicitly define ownership as claims on residual cash flows. In Grossman-Hart-Moore’s perspective shareholders are the “owners” because they have the voting power to determine how assets are deployed, whereas in Berle and Means shareholders are the “owners” because they have the rights to residual cash flows. Thus, ownership of unique non-human assets is, perhaps, the primary source of power in the corporation. 36 That is having the right to make any decisions concerning the asset’s use that are not explicitly controlled by law or contractually assigned to another part (see, e.g., Grossman and Hart 1986). 37 A plausible explanation for this difficulty is the bounded rationality faced by economic agents in their decision-making. Williamson (1990) makes the point that in a bounded rationality setting “all complex contracts are unavoidably incomplete.” 38 One that specifies what each party is to do in every relevant eventuality at every future date and how the resulting income in each such event should be divided. Thus, an arrangement with the ability to explicit procedures governing the behavior of contract participants in determining outcomes as well as the allocations resulting from those outcomes. 39 According to Hart (1988) Ronald Coase and Oliver Williamson should be credited for the «insight that the firm as an institution takes on importance only in a world of incomplete contracts.» Tirole (1999) provides a comprehensive and rigorous account of the incomplete contracts literature. 40 Assets can be relationship-specific creating a potential for costly hold-up problems to emerge (e.g., Klein, Crawford and Alchian 1978; and Williamson 1975). Hart and Moore (1988) hypothesize that hold-up problems can create underinvestment. 41 Allocation of rents in financial contracts is in general (endogenously) contingent on investment policy and operating decision-making. Harris and Raviv (1992) observe that the design of financial contracts also includes the «assignment of control rights.» 42 Property rights theory has little to say about debt contracts, and/or the relative importance of debt and equity, since ownership rights are, by definition, vested in the owners of the firm. Differences in control rights in distinct states of world are also ignored.

- 11 -

nature of control rights.43 Through the incomplete contract “lens”, standard financing contracts,

either debt or equity arrangements, are viewed as conferring rights to both control over

managerial decision-making and cash flow streams.44

As it is widely accepted, in a Modigliani and Miller’s (1958) economy, where transaction

costs are assumed away, and perfect and costless observability of agent’s actions is possible to

design and write an ex ante incentive contract to induce the agent to act in the principal’s best

interest.45 Furthermore, these contracts could be costlessly executed and enforced.46 If a binding

contract laying down each party’s obligation and payoffs for any conceivable eventuality in

every possible future state of the world could be written at no cost, the so-called agency

problems would not emerge.47 Hence, agency problems seem to be associated with the imperfect

observability of agent’s actions and the costs of writing, executing, and enforcing contracts.

It is also widely accepted that, only under a very restrictive, strong and well-known set of

circumstances firm’s insiders and outside investors fully share all available information,

including the distribution of returns accruing to real investment opportunities. In this setting,

demand- and supply-side capital market participants agree about the value of alternative

financing plans, and equilibrium security prices make the firm indifferent among alternative

security issues.

The most conspicuous agency problems related to capital structure choice are commonly

associated with incentive conflicts stemming from (incomplete) contracting between

shareholders and managers, and between shareholders and debtholders.48 These problems

recognizably propel potential inefficiency, and adverse wealth effects, which can affect

investors’ valuation of firm’s securities, and even their willingness to supply funds.

43 I.e., contractual stipulations make the allocation of control dependent on a future state of the world. In a complete contracts economy ex post allocation of control rights is clearly void of meaning. 44 According to Berglöf (1990) «a firm’s capital structure can be viewed as describing the allocation of risk and control among investors.» Garvey and Hanka (1999) suggest that «corporate managers have discretion over capital structure choices, as the firm’s founding shareholders cannot write a comprehensive ex ante contract specifying all future financing decisions.» 45 It is equivalent to writing a complete contract. See, e.g., Tirole (1999) for further details and references to the related literature. 46 However, as pointed out by the incomplete contract literature, costs incurred in foreseeing future contingencies, and in writing and enforcing contracts are non-trivial (e.g., Tirole 1999, 743-4). 47 Agency theory attempts to explain the principal-agent relationship using the metaphor of a contract (Eisenhardt 1989) in which one party, the principal, delegates work to another party, the agent, who is empowered with some decision-making power in order to perform that work (e.g., Jensen and Meckling 1976). 48 Agency theory is also useful to explain conflicts of interests emerging within agency relationships between headquarters’ managers and divisional managers in internal capital market settings (see, e.g., Bolton and Scharfstein 1998).

- 12 -

However, the weaknesses of the neo-classical paradigm go beyond its apparent inability

to incorporate important characteristics of owners and decision-makers’ contracting behavior in

real-world firms. Most notably, it also seems to ignore the role ascribed to investment in a firm’s

securities, which are viewed as pure financial assets void of any underlying power of economic

decision which, autrement dit, it is not viewed as an economic good.49 Summarizing, it seems that if we want to build a corporate framework able to explain

firms’ financing behavior, we must incorporate the notion that an equity instrument grants

property rights over the actions of governance of the firm, and establishes a mechanism for the

transfer of control.50 A clear focus on the so-called contractual theories of the firm is needed to

account for phenomena of property rights, asymmetric information, and self-interest behavior.

The combined notions of property rights (and therefore incentives), asymmetric information and

extended self-interest appear to be a sufficiently rich toolbox to potentially enable the

rationalization of the capital structure problem in the context of a firm’s financing behavior.

Moreover, understanding the role of the debt/equity choice is not a simple matter, because

financial markets are notoriously frictional and incomplete, and financing contractual

arrangements are unavoidably incomplete. Whenever these conditions hold, every profitable

investment opportunity is funded at cost commensurate with its risk class, regardless of the way

cash flow claims are divided among claimholders. Therefore, we need to depart from M&M

(1958) ideal world if we envision shedding some light on Dowd’s (1996, 601) question «Why do

agents use the particular contract forms debt and equity contracts in particular that we observe in the ‘real

world’?»

2.1. CAPITAL STRUCTURE POLICY MODELS

If capital structure definition and readjustment do matter, it follows then such decisions

are not random as implied by M&M (1958), and that decision-makers at the firm level may

arguably adopt a policy model to guide their choices. Kraus and Litzenberger (1973), Scott

(1976), Kim (1978) and Chen (1978, 1979), among others, explain that firms choose their mix of

debt and equity financing by trading-off expected costs and benefits of debt financing. The

theory describes a firm’s optimal capital structure as the mix of financing that equates the

49 According to Milgrom and Roberts (1992) «the classical theory regards financial securities as claims on streams of net receipts whose magnitude and variability are exogenously given.» Berglöf (1990) points out that «in modern finance literature à la Modigliani and Miller (1958) �…� financial instruments only entitle their holders to return streams.» 50 The notion that control rights are residual builds on the fact that control rights holders are constrained, among other things, by law and other parties’ contractual rights, in the array of actions they actually can undertake.

- 13 -

marginal costs and benefits of debt financing. A major empirical prediction of the tradeoff theory

is that debt ratios will tend to be mean reverting, as firms use external capital markets to keep

themselves at or close to their optimum debt levels. Trade-off theories, however, leave

undetermined its underlying economic framework, and unsolved the identification and

measurement of costs and benefits.

Among the benefits included in the trade-off equation are the tax advantages of debt

financing (Modigliani and Miller 1963), the benefits associated with the control of free cash flow

problems (Jensen 1986), and the disciplinary effects of leveraging up over managerial discretion

(Jensen and Meckling 1976). Among the costs are the costs of financial distress, and agency

costs of debt and equity financing (Jensen and Meckling 1976; Myers 1977; Stulz 1990; and Hart

and Moore 1995). Ross (1977) models the costs and benefits of signaling with capital structure.

According to Myers (1984) and Myers and Majluf (1984) corporate financing choices are

driven by the costs of adverse selection arising as a result of information asymmetry between

better informed managers and less informed investors. Since these costs are incurred only when

firms issue securities and are lower for debt than for equity, firms should prefer internal

financing and prefer debt to equity when external funds have to be raised. Building on early

work by Donaldson (1961), those authors developed the pecking order theory of capital

structure. The major prediction of the model is that firms will not have a target or optimal capital

structure, but will instead finance new investment opportunities following a pecking order in

using and exhausting available financing sources: available financial slack and internally

generated cash flows first, followed by new debt issues, and finally, only when the firm reaches

its ‘debt capacity’, new equity financing.

The neutral mutation theory proposed by Miller (1977) suggests that firms may develop

financing habits that tend to «follow the lines of least resistance as well as least damage».( Kamath 1997).

The market timing theory developed by Baker and Wurgler (2001), is an empirically motivated

hypothesis based on strong statistical evidence that financial leverage is inversely related to a

measure of historical market valuation. The theory motivates the prediction that firm stock price

performance may be relevant for timing firm’s new security offerings.

- 14 -

2.2. DETERMINANTS OF CAPITAL STRUCTURE POLICY

The more conspicuous theoretical and empirically motivated capital structure

determinants are related to: (1) the lack of neutrality in the taxation regime with respect to

financing costs and cash flow distributions to claimholders;51 (2) principal-agent conflicts of

interest and governance; (3) differently informed contracting parties; and (4) behavior on product

/ input markets.

Under specific circumstances, the differential tax treatment of equity and debt securities

tends to generate an incentive for a preference for debt financing. Thus, if effective tax rates are

exogenous and common at the firm and the individual levels, firms will tend to resort to the most

tax-advantaged source of financing. Firm behavior in this respect should to be closely related to

the idiosyncratic characteristics of the tax regime. The main hypotheses in this heading are: (1)

the tax advantage of borrowing costs (Modigliani and Miller 1963); (2) investors’ income

taxation (Miller 1977); and (3) non-debt tax shields such as depreciation, provisions and tax

carryovers (DeAngelo and Masulis 1980).52

In 1963, Modigliani and Miller corrected their 1958 paper waiving the presumption of a

taxless economy, and deriving a corner solution for the firm’s optimal capital structure problem,

which leads to an unambiguous (almost) infinite debt-to-equity ratio. The kernel of M-M’s

reasoning is that, by making interest expense tax deductible, the government is subsidizing firms

that finance their operations and projects by issuing debt securities. Therefore, ceteris paribus,

firms would try to maximize their share of the government subsidy when choosing their capital

structure. Thus, when income-tax deductibility of interest payments at the firm level is present,

the market value of the firm is an increasing function of its financial leverage. Therefore, the

value of the firm is maximized with an all-debt capital structure.53

51 According to Faig and Shum (1999) an «asymmetric corporate tax system is one in which the government does not treat firms' gains and losses equally» The decision of US government banning interest deductibility on very-long-term bonds (maturities of 40 years and over) and reclassifying them, for tax purposes, as equity is an illustration of reducing tax asymmetries between debt and equity securities 52 See also Bradley, Jarrell, and Kim (1984), Titman and Wessels (1988), MacKie-Mason (1990), Givoly, Hahn, Ofer, and Sarig (1992), Gentry (1994), Graham (1996), and Graham, Lemmon, and Schallheim (1998) for studies on the relationship between taxes and firm’s capital structure. 53 According to Modigliani (1988), M&M (1963) argument rests on the assumption that the tax saving stream «is constant, perpetual, and absolutely certain like the coupon of a government bond.»

- 15 -

Principal-agent problems create potential conflicts of interest over the appropriation of

private control rents, the manifestation of divergences over property rights, and the propensity

for opportunistic behavior. In a homogeneous expectations economy it would be possible, ex

ante, to write incentive contracts to induce agents to act in the principal’s best interest. .

As argued by Jensen and Meckling (1976), firm’s capital structure choice affects

concurrently managerial and providers of capital incentives.54 Not surprisingly, most

conspicuous agency problems relating to firm’s financing structure are though associated with

incentives stemming from (incomplete) contractual arrangements among shareholders and

managers, and shareholders and debtholders. These problems induce adverse welfare effects. In

these instances, the principal will attempt to limit the extent of the misalignment over objective

functions, by contracting the disputes over agent’s incentives to engage in potentially wealth

harmful activities. Such appropriate binding incentives for the agent are costly, and therefore

make the principal to incur in transaction costs of different nature.

Whenever the access to complete and perfect information is restricted and costly, parties

in a binding contractual arrangement tend to be unevenly informed. This asymmetry of

information creates an incentive for the superiorly informed party to behave opportunistically,

either by hiding its true characteristics (adverse selection) or hiding its post-contractual actions

(moral hazard). Ultimately, imperfect and imperfect costly information fosters uncertainty and

therefore creates the incentive for opportunistic behavior.

Imperfect, incomplete and costly information about the ‘quality’ (cash flow riskiness) of

borrowers’ investment projects creates an asymmetric information problem between borrowers

and lenders. Consequently, adverse selection leads, in an unevenly informed capital market, to a

gap between the cost of external financing and internal financing (‘lemons’ premium). In the

presence of incentive problems and costly monitoring of managerial actions, rational

expectations suppliers of external financing require a higher return to compensate them for: (1)

monitoring costs; and (2) moral hazard costs associated with managerial discretion over resource

allocation.

Whenever a firm has private information about its prospects, its securities may be

mispriced (Ackerlof 1970). In this framework, firm’s insiders may convey information to

54 The incomplete contracts approach, for example, recognizes that financiers have differential incentives regarding the choice of project risk and that this choice should optimally be contingent on past firm performance (Aghion and Bolton, 1992, Dewatripont and Tirole, 1994, Hart, 1995).

- 16 -

uninformed market participants through discretionary external financing decisions to signal their

true characteristics, reduce the informational gap and therefore minimize adverse selection costs.

A firm’s financing structure may affect its incentives to produce and invest (Harris and

Raviv 1991). Therefore it may determine strategic interactions between the firm’s capital

structure choice and the behavior vis-a-vis its customers, suppliers and competitors. Since the use

of debt financing can convey some economic profits, firms with less ‘plastic assets’ could have

higher financial leveraged than firms with more ‘plastic assets’ (Alchian and Woodward 1987,

1988).

Firms that enter in long-term contractual relationships with their customers, create an

implicit expectation of sustained flow of supply non-trivial switching costs. A firm providing

such type of goods or services may incur in a demand decline for their products if it leverages up

its capital structures and such behavior is perceived by customers as potentially harmful for the

firm’s likelihood of bankruptcy. Ceteris paribus, we should expect that such firms would carry

less debt than producers of nondurable goods (Titman (1984).

Firms with larger investments in reputation to guarantee product quality will employ

lower financial leverage. The increased incentive of equityholders to depreciate quality resulting

from increased leverage causes a decline in the value of reputation. Therefore we should expect

firms with large reputation capital to use lower level of debt financing (Chung and Smith 1987).

The literature abundantly documents empirical regularities of (non-financial) firms’

capital structures.55 The most ubiquitous results include: (1) the presence of a persistent industry

effect (e.g., Damodaran 1997; and Hanjoon Kim 1999); (2) similarities in choice of financial

leverage at the firm level (e.g., Bowen, Daley and Huber 1982; and Boquist and Moore 1984);

(3) positive correlation between leverage at the firm and the industry level. (e.g., Castanias 1983;

and Bradley, Jarrell and Kim 1984); and (4) firm’s characteristics such as: size, R&D intensity,

market-to-book, stock returns, asset tangibility, profitability, and marginal tax rate, as important

determinants of capital structure (e.g., Jalilvand and Harris 1984; Titman and Wessels 1988;

Bayless and Chaplinsky 1990; MacKie-Mason 1990; Rajan and Zingales 1995; Graham 1996;

Jung, Kim, and Stulz 1996). An overall assessment of capital structure empirical literature seems

to indicate: (1) an ambiguous validation of extant theories; (2) the presence of inconsistencies

55 For reviews of this literature see, among others, Masulis (1988), Harris and Raviv (1981), and Mario Santos (2000).

- 17 -

and contradictions between stylized facts and theoretical predictions; and (3) that empirical

results appear strongly dependent of samples and research methods.

2.3. CAPITAL STRUCTURE OF THE BANKING FIRM

Banks are simultaneously firms, financial intermediaries and regulated entities. The

nature of the incentives induced by regulatory jurisdiction determines a unique interaction

between bank’s capital base and its behavior.

Given the mandatory requirements for banks’ capital standards, banking firms are

involved in both voluntary and involuntary capital structure decisions (e.g., Besanko and Kanatas

1996; Cornett and Tehranian 1994; and Keeley 1989). Voluntary capital structure decisions are

taken in the very same setting as non-financial firms and, arguably, under the same determinants

that are hypothesized in the capital structure theory for those firms. Involuntary capital structure

decisions are enforced by compliance prescriptions dictated by violations of the mandatory

regime of capital adequacy requirements issued by regulators.56 Although recognizing their

importance and interaction, the focus of this study is confined to the voluntary capital structure

decisions.

3. RESEARCH DESIGN, METHODOLOGY, AND EMPIRICAL IMPLEMENTATION

We build on the general theory of capital structure, typically associated with the non-

financial firm, to develop a theoretical framework able to support the formulation of testable

propositions included in the survey instrument.

The investigation is primarily concerned with establishing an empirical link between

testable propositions in the form of survey questions, and capital structure theory of the banking

firm. More specifically, we use survey scores of a sample of banks’ Chief Executive Officers

(CEOs)57 spaning the 1989-1998 decade, to document preferences for capital structure policy

models, and to uncover the empirical relevance of determinants of Portuguese banks’ voluntary

capital structure choice.

The survey design was developed taking in consideration: (1) the inferential nature of the

investigation; (2) the concern with minimization of the non-response bias, the potential for

response-bias, and to avoid sampling bias; (3) the intrinsic technicality of the topic; (4)

56 For a recent survey of theoretical literature on banks capital regulation see João C. Santos (2000). 57 We adopted the CEO/bank as the unit of analysis of this research.

- 18 -

respondents’ required level of understanding of the survey questions; and (5) to achieve a high

response rate.58, 59

The target population was defined as the CEOs of all the banks incorporated as

autonomous entities under the Portuguese law with tenure of, at least two years, during the

period 1989 to 1998.60 The survey was administered to a sample of 51 CEOs representing 89.5

percent of the population, which compares with an average 20.3 percent response rate in capital

structure mail-administered surveys. Six CEOs were unavailable or unwilling to participate in

the survey [Insert Table 1].

The following criteria were adopted to define the target population: First, we excluded

the CEOs of all foreign banks (we were very skeptical about the possibility of having the

opportunity to conduct personal interviews to the CEOs of those international banks).61 Second,

CEOs whose tenure was less than an arbitrarily set limit of two years were not included in the

target population, because we view the horizon of capital structure planning going beyond that

time limit. Third, we adopted the view that CEOs of banks that at 1998 year-end were

incorporated for less than two years were also not eligible as members of the population. Here

the rationale is related to the fact that authorities require incorporating banks to hold a minimum

amount of capital. We assumed that such amount of capital was likely to be adequate for the first

two years and therefore no capital structure decisions were likely to be taken during that period

of time. Four, CEOs of mutual banks were not included because of the dissimilarities between

the objective function of these financial intermediaries when compared to banks organized as

open-investment corporations. In the Portuguese banking system there is only one ‘true’ mutual

bank.

Our work distinguishes from prior research using the same methodological approach, in a

number of aspects. First, our survey was conducted in a face-to-face interview format rather than

administered by mail, as is the case with the majority of studies, in sharp contrast with the great

58 Validity of findings is significantly improved in a face-to-face interview survey format because, among other reasons, the interviewer as the opportunity to «clarify the meanings of questions and response choices» (Schober and Conrad 1997) and to homogenize perception of survey participants about formulated questions. 59 These are debilities often associated with survey-based research, mainly in mail-administered format. 60 To identify the population we conducted a documental analysis of the composition of banks’ management teams which developed through annual reports we collected through a mailing to all banks. We constructed a database of bank’s Board of Directors or body of governance performing similar functions and having the same accountability profile. Based on that data we built the listing of bank’s CEOs. 61 Banks whose controlling shareholders are non-residents have their capital structure decisions, most likely, taken at international level.

- 19 -

majority of capital structure survey-based research.62, 63 Since the interviews were personally

conducted by one of the authors, it might have also contributed to enhance the likelihood of

reducing the variance in CEOs’ interpretation of survey questions. Secondly, the study was

conducted within a single and relatively homogeneous industry, thus avoiding difficulties in

controlling the unsystematic effects, inevitably present in cross-sectional samples. Unlike most

recent survey-based studies of capital structure (e.g. Fan and So 2000; Graham and Harvey 2001;

and Bancel and Mittou 2002), our investigation focus on a single industry rather than surveying a

broad set of industries. Therefore, we expect that might have helped in circumventing the classic

difficulties of controlling the unsystematic effects and in reducing the variance inevitably

associated with idiosyncratic chocks and features of that nature inevitability present in each of

the industries included in cross-sectional data. Third, although confidentiality and anonymity

were guaranteed to survey participants, we were able to control banks characteristics. Fourth, the

survey was conducted outside the U.S. and consequently offered a new opportunity for

increasing the generalization power of the theoretical propositions submitted to empirical testing.

Lastly, we were able to simultaneously examine qualitative data obtained from our survey and

quantitative data drawn from our database.

The period 1989-1998 represented a challenge for the research. During that time, a

number of events unfolded at the Portuguese bank level: privatizations, hostile takeovers,

mergers and other forms of restructuring, and internationalization. At the macroeconomic level,

Portugal faced the effects of increasing integration of economies and markets, the global

deregulatory trend and restructuring pressures dictated by its admission to the European

Economic Community in 1986.

A fifty five closed questions questionnaire was used including: (1) descriptive questions,

to elucidate relevant contextual aspects of banks’ capital structure choice; (2) conceptual /

theoretical questions, testable propositions; and control questions.64A pre-testing was conducted

using a preliminary version of the questionnaire.

62 For an account of capital structure survey-based literature see Mario Santos (2000). 63 The survey instrument is available from the authors. 64 A six points Likert scale was used in order to force respondents to discriminate, either positively or negatively, minimizing the tendency for mean answering behavior, typically associated with odd Likert scales.

- 20 -

4. RESULTS

In order to deepen our analysis, we split our sample by: (1) ‘State-Owned’ and ‘Privately

Owned’ Bank CEOs; (2) ‘Listed’ and ‘Unlisted’ Bank CEOs; (3) ‘DeNovo’ and ‘Established’

Bank CEOs; and (4) ‘Over-Capitalized’ and ‘Under-Capitalized’ Bank CEOs. We build this last

binary variable by classifying as ‘over capitalized’ the banks whose average capital ratio during

the CEO tenure was higher than the industry’s average capital ratio during the same time period.

We classified as ‘under capitalized’ those banks whose average capital ratio during the CEO

tenure was lower than the industry’s average capital ratio.65

Findings suggest that capital structure decisions at the bank level do matter [Insert Tables

1, 2, and 3]. Not surprisingly, the evidence supports the view that managerial strategic financing

decision-making is not random, as implied by M&M’s irrelevance hypothesis. Responses of

survey participants show their concern for the implications of capital structure decisions on bank

valuation.

Results show moderate support for the trade-off theories of capital structure. Little

evidence was found in favor of the pecking order capital structure policy model [Insert Table 4].

54.9 percent of surveyed CEOs admit that bank’s stock price performance is relevant for timing

new security offerings. This evidence is interpreted as consistent with the market-timing theory.

Overall, changes in regulatory apparatus were deemed as the major external factor

affecting CEOs capital structure decisions. Takeover threats are not perceived as a meaningful

corporate control disciplinary device, probably because of regulatory intervention in the market

for corporate control. Capital market performance is a statistically significant factor for privately

owned Bank CEOs, as well as for listed bank CEOs, although not statistically significant [Insert

Tables 6 and 6a].

Ownership structure and managerial control, investment policy / growth opportunities,

financing flexibility, and bank’s reputation in credit and deposit markets were the relevant

internal determinants of capital structure decisions indicated by survey participants. A similar

pattern is observed when we split the sample [Insert Tables 7 and 7a].

65 We used data from our banks’ financial statement database to compute each bank weighted (by deflated net total assets) average capital ratio during the tenure of each CEO included in our sample. We also calculated the weighted (by the deflated value of net total assets) average capital ratio for all the banks in sample during the same tenure period, and used it as a proxy for the target capital structure.

- 21 -

Other results show: (1) no evidence consistent with the view that banks’ insiders might

systematically engage in various forms of excessive risk-taking. Thus, the moral hazard

incentives associated with the deposit insurance regime (and other forms of safety net

guarantees) might be less severe in Portugal than in other countries, such as the U.S.;66 (2) some

support for the disciplinary role of debt, capital markets and regulatory jurisdiction. However,

the strong support received by effect of the too big to fail doctrine on capital structure suggests

that managerial risk aversion induced by impending bankruptcy might be small; (3) Portuguese

banks might be less exposed to severe undervaluation caused by adverse selection problems.

Reported financial flexibility is consistent with this interpretation; (4) product markets and

market for corporate control were not found to be effective mechanisms to mitigate conflicts

arising within agency arrangements, and thus to discipline insiders; (5) taxation at the bank level

as well as bankruptcy and financial distress considerations were found to be only moderately

influential on banks’ capital structure; (6) taxation at the personal level, transactions costs, free

cash flow considerations, were not found to unambiguously affect banks’ capital structure

choice; (7) signaling, underinvestment and asset substitution problems were found to moderately

affect the debt / equity choice.

5. CONCLUSIONS

Results allow the establishment of an empirical link between capital structure theory and

debt / equity choice of surveyed CEOs. The managerial perspective embodied in the survey

suggests that capital structure policy seems to be more affected by incentives structure and

governance control rights underlying the different financing instruments, rather than by the

aspects related to security design and pricing. We interpret this finding as providing support to

the notion that theories which elucidate the debt-equity choice in non-financial firms, seem

likewise adequate to explain banks capital structure decisions.

It was documented that capital structure choice, as we hypothesized, shows some

(varying) consistency with a number of theoretical propositions. From this we derive the

implication that the theories which explain the debt-equity choice in non-financial firms seem

also able to accommodate banks’ capital structure decisions, once their financial intermediation

idiosyncrasies are taken into consideration [Insert Table 8].

66 Additional explanations may be related to the paucity of “problematic” banks in Portugal, as well as to differences in Portuguese deposit insurance regime compared to the U.S.

- 22 -

Prevailing ownership structure among Portuguese banks is distant from the prototypical

Berle and Means diffusely and publicly held firm that usually underlies a number of capital

structure theories. Informational and governance consequences of this fact had to be properly

taken into consideration when drawing conclusions from our survey data. Ownership structure,

and therefore its informational and governance consequences, appears to be a material element

for the understanding of capital structure behavior.

The intuition that capital structure choice is likely to assume the form of a trade-off

between costs and benefits associated with such a decision is a well-known useful and appealing

concept. Unfortunately, it is manifestly clear that it cannot resolve the central problem of

identifying and measuring these costs and benefits, thus leaving undetermined the economic

framework that could explain the capital structure conundrum. We believe that there is nothing

wrong with the trade-off approach. We also recognize that very important steps have been made

in improving our understanding about the influence of behavioral considerations like

incentives and governance arrangements of capital structure decisions. The acknowledgement

that people, not production functions (or any invisible hand), actually make these choices was a

significant contribution we owe to Michael Jensen and Bill Meckling, among others. The

recognition that there are ownership rights embedded in the securities sold by firms to manage

their capital structures was another important step.67 Yet another was the perception that buyers

and sellers of securities typically get separated and that this could affect the costless exercise of

ownership rights. Finally, the acceptance that individuals, in making their choices (firms are just

real fictions…they do not make decisions!), are unable to behave according to the full rationality

paradigm, further extended our understanding of economic behavior. This accumulated

knowledge, however, it is not enough to enable the construction of a comprehensive capital

structure theory. In this framework, capital structure decisions should be primarily determined by

considerations that relate to incentives and allocation of ownership and control rights. Given that

managerial reputation is a central factor in motivating risk-averse corporate managers (Fama and

Jensen 1983), we should expect their behavior to be also affected by problems with their human

capital specific investments.

67 Central to this evolution was the residual rights of control concept introduced by Grossman and Hart (1986). Williamson (1988) argues that "rather than regard debt and equity as 'financial instruments', they are better regarded as different governance structures."

- 23 -

In our assessment, the empirical findings obtained from our survey are consistent with

this perspective. Managers appear to be primarily concerned with the influence of the incentives

associated with the governance arrangements and the control rights allocation determined by

capital structure decision-making.68 Their responses seem to indicate less concern with security

design, transaction costs and other tactical aspects of the capital structure problem.

This is consistent with Stewart Myers’ viewpoint69 that one possible explanation for the

capital structure puzzle might be related with the excessive emphasis put on financing structure

(he called the tactical level of capital structure), in detriment of financial structure (he called the

strategic level of capital structure).70

Two concluding remarks. The first to acknowledge that we lack a theory with the ability

to explain and predict the dynamics of a firm’s capital structure choice along its life cycle.71 This

theory should be able to enlighten firm’s debt / equity choice such as the decisions to go public

and the decisions to go private. The dynamics, over the life cycle of the firm, of crucial factors

for capital structure choice determine a complex set of interactions from where it has been

difficult to disentangle the individual components. Among those elements is the structure of

managerial incentives and therefore managers’ motivations and expectations, the organizational

form of the firm, as well as its ownership structure. The second relates to restrictions to pure

capital structure decisions that are present when share repurchases are restricted, as it is the case

in Portugal. We consider this a public policy issue that should deserve, at least, reconsideration at

the European Union legislative level. European legislators emphasize creditor protection to

restrict stock buybacks. However, an argument can be made that banks’ largest creditors are

depositors whose claims typically benefit from the protection of the public deposit insurer. Other

banks’ debtholders are usually well informed and sophisticated investors whom are able to

accurately and efficiently appraise and price default risk. Furthermore, bank debt offerings are

often made under private placement arrangements and “sweetened” by rating notations.

Moreover, bank insiders if deprived of an effective defense against hostile takeover threats might

68 The responses to the motivation to issue convertible securities illustrate this point. 69 See the 1998 Vanderbuilt University “Roundtable on The Capital Structure Puzzle”, and Myers (1999). 70 The first of these two aspects is related to an operating view of finance in the sense that the kernel of its object is, essentially, an attempt to overcome financial markets incompleteness (we espouse the view that the system of financial markets is not complete) through innovative strategies of security design and pricing. See, e.g. Allen and Gale (1994), Harris and Raviv (1992) and Ross (1989) among others, provide a thoughtful review of this literature. 71 According to Myers (2001), “there is no universal theory of the debt-equity choice, and no reason to expect one.”

- 24 -

resort to external control mechanisms (such as anti-takeover charter amendments) and insulate

themselves from the discipline of the market for corporate control.

- 25 -

REFERENCES Alchian, Armen and Harold Demsetz. 1972. Production, Information Costs, and Economic Organization.

American Economic Review 62(5): 777-795. Allen, Franklin and Andrew Winston. 1995. Corporate Financial Structure, Incentives and Optimal

Contracting. In Finance - Handbooks in Operations Research and Management Science, vol. 9, ed. R. A. Jarrow, V. Maksimovic, and W. T. Ziemba, 693-720. Amsterdam: Elsevier.

Allen, Franklin and Douglas Gale. 1994. Optimal Security Design. In Financial Innovation and Risk Sharing, ed. F. Allen, and D. Gale, chapter 6. Cambridge (MA): MIT Press.

Ang, James, Rebel Cole and James Wuh Lin. 2000. Agency Costs and Ownership Structure. Journal of Finance 55(1): 81-106.

Auerbach, Alan. 1992. Leverage. In The New Palgrave Dictionary of Money and Finance, Vol.2, ed. P. Newman, M. Milgate, and J. Eatwell, 574-575. New York (NY): Macmillan Press.

Baker, Malcolm and Jeffrey Wurgler. 2002. Market Timing and Capital Structure. Journal of Finance 57(1): 1-32.

Bancel, Franck and Usha R. Mittoo. 2002. The Determinants of Capital Structure Choice: A Survey of European Firms. Working paper presented to the 2003 Annual Meeting of the American Finance Association, Washington (D.C).

Baron, David. 1974. Default Risk, Homemade Leverage, and the Modigliani-Miller Theorem. American Economic Review 64(1): 176-182.

Baron, David. 1974. Default Risk, Homemade Leverage, and the Modigliani-Miller Theorem. American Economic Review 64(1): 176-182.

Baron, David. 1976. Default Risk and the Modigliani-Miller Theorem: A Synthesis. American Economic Review 66(1): 204-212.

Baron, David. 1976. Default Risk and the Modigliani-Miller Theorem: A Synthesis. American Economic Review 66(1): 204-212.

Berger, Allen and Gregory Udell. 1998. The Economics of Small Business Finance: The Roles of Private Equity and Debt Markets in the Financial Growth Cycle. Journal of Banking and Finance 22(6-8): 613-673.

Berglöf, Erik. 1990. Capital Structure as a Mechanism of Control: a Comparison of Financial Systems. In The Firm as a Nexus of Treaties, ed. M. Aoki, B. Gustafsson, and O. E. Williamson, 237-262. London: Sage Publications.

Berle, Adolph and Gardiner Means. 1932. The Modern Corporation and Private Property. New York (NY): Macmillan Publishing Co.

Bhabra, Harjeet, Tong Liu and Dogan Tirtiroglu. 2002. Capital Structure Choice in a Nascent Market: Evidence from Listed Firms in China. Working paper presented to the 2002 Annual Meeting of the Financial Management Association International, San Antonio (TX).

Bolton, Patrick and David Scharfstein. 1998. Corporate Finance, the Theory of the Firm, and Organizations. Journal of Economic Perspectives 12(4): 95-114.

Booth, Laurence, Varouj Aivazian, Asli Demirgüç-Kunt and Vojislav Maksimovic. 2001. Capital Structures in Developing Countries. Journal of Finance 56(1): 87-130.

Chen, Andrew and Sumon Mazumdar. 1994. Impact of Regulatory Interactions on Bank Capital Structure. Journal of Financial Services Research 8(4): 283-300.

Chen, Andrew. 1979. A Theory of Corporate Bankruptcy and Optimal Capital Structure. In Handbook of Financial Economics, ed. J. L. Bicksler, 275-291. Amsterdam: North-Holland.

Demirgüç-Kunt, Asli and Vojislav Maksimovic. 1996. Stock Market Development and Corporate Finance Decisions. Finance & Development 33(2): 47-49.

Donaldson, Gordon. 1961. Corporate Debt Capacity: A Study of Corporate Debt Policy and the Determination of Corporate Debt Capacity. Cambridge: Graduate School of Business Administration, Division of Research, Harvard University.

- 26 -

Dowd, Kevin. 1996. Competition and Finance: a Reinterpretation of Financial and Monetary Economics. New York: St. Martin’s Press.

Easterbrook, Frank and Daniel Fischel. 1991. The Economic Structure of Corporate Law. Cambridge (MA): Harvard University Press.

Eggertsson, Thráinn. 1990. Economic Behavior and Institutions. Cambridge (UK): Cambridge University Press.

Eisenhardt, Kathleen. 1989. Agency Theory: An Assessment and Review. Academy of Management Review 14(1): 57-74.

Fama, Eugene and Merton Miller. 1972. The Theory of Finance. Hinsdale: Dryden Press. Fama, Eugene and Michael Jensen. 1983a. Agency Problems and Residual Claims. Journal of Law and

Economics 26: 327-349. Fama, Eugene and Michael Jensen. 1983b. Separation of Ownership and Control. Journal of Law and

Economics 26(June): 301-325. Fama, Eugene. 1980. Agency Problems and the Theory of the Firm. Journal of Political Economy 88:

288-307. Fan, Dennis and Raymond So, 2000. A Survey on Capital Structure Decisions of Hong Kong Firms.

Review of Pacific Basin Financial Markets and Policies 3(3): 347-365. Fisher, Edwin, Robert Heinkel and Josef Zechner. 1989. Dynamic Capital Structure Choice: Theory and

Tests. Journal of Finance 44(1): 19-40. Frankfurter, George and George Philippatos. 1992. Financial Theory and the Growth of Scientific

Knowledge: From Modigliani and Miller to an "Organizational Theory of Capital Structure". International Review of Financial Analysis 1(1): 1-15.

Froot, Kenneth. 1995. Incentive Problems in Financial Contracting: Impacts on Corporate Financing, Investment and Risk Management Policies. In The Global Financial System: A Functional Perspective. ed. Crane, Froot, Mason, Perold, Merton, Bodie, Sirri and Tufano, 225-261. Boston (MA): Harvard Business School Press.

Garvey, Gerald and Gordon Hanka. 1999. Capital Structure and Corporate Control: The Effect of Antitakeover Statutes on Firm Leverage. Journal of Finance 54(2): 519-546.

Gertler, Mark. 1988. Financial Structure and Aggregate Economic Activity: An Overview. Journal of Money, Credit and Banking 20 (August, part 2): 559-588.

Graham, John and Campbell Harvey. 2001. The Theory and Practice of Corporate Finance: Evidence from the Field. Forthcoming Journal of Financial Economics 60(2-3): 187-243.

Grossman, Sanford and Oliver Hart. 1982. Corporate Financial Structure and Managerial Incentives. In The Economics of Information and Uncertainty, ed. J. McCall, 107-140. Chicago (IL): University of Chicago Press.

Grossman, Sanford and Oliver Hart.1986. The Costs and the Benefits of Ownership: A Theory of Vertical and Lateral Integration. Journal of Political Economy 94: 691-719.

Hagen, Kåre. 1976. Default Risk, Homemade Leverage, and the Modigliani-Miller Theorem: Note. American Economic Review 66(1): 199-203.

Hansmann, Henry. 1988. Ownership of the Firm. Journal of Law and Organization 4(2): 267-304. Hansmann, Henry. 1996. The Ownership of Enterprise. Cambridge (MA): The Belknap Press of Harvard

University Press. Harris, Milton and Arthur Raviv. 1991. The Theory of Capital Structure. Journal of Finance 46(1): 297-

355. Harris, Milton and Artur Raviv. 1988. Corporate Control Contests and Capital Structure. Journal of

Financial Economics 20: 55-86. Harris, Milton and Artur Raviv. 1992. Financial Contracting Theory. In Advances in Economic Theory

Sixth World Congress, Volume II, ed. Jean-Jacques Laffont, 64-150. Cambridge: Cambridge University Press.

Hart, Oliver and John Moore. 1988. Incomplete Contracts and Renegotiation. Econometrica 56(4): 755-785.

- 27 -

Hart, Oliver and John Moore. 1990. Property Rights and the Nature of Firms. Journal of Political Economy 98:1119-58.

Hart, Oliver, and John Moore. 1995. Debt and Seniority: An Analysis of the Role of Hard Claims in Constraining Management. American Economic Review 85: 567-585.

Hart, Oliver. 1988. Capital Structure as a Control Mechanism in Corporations. Canadian Journal of Economics 21(3): 467-476.

Hasan, Muhammad. 1997. The Determinants of Capital Structure of Banks. Kent State University Graduate School of Management Ph.D Dissertation. UMI number 9809246. Ann Arbor: UMI.

Hellwig, Martin. 1981. Bankruptcy, Limited Liability, and the Modigliani-Miller Theorem. American Economic Review 71(1): 155-170.

Hovakimian, Armen, Tim Opler and Sheridan Titman. 2001. The Debt-Equity Choice. Journal of Financial and Quantitative Analysis 36(1): 1-24.

Jensen, Michael and Clifford Smith. 1985. Stockholder, Manager, and Creditor Interests: Applications of Agency Theory. In Recent Advances in Corporate Finance, ed. E. Altman and M. Subrahmanyam, 93-131. Homewood (IL): Richard D. Irwin.

Jensen, Michael and William Meckling. 1976. Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure. Journal of Financial Economics 3(4): 305-360. Revised edition, April 1997. Mjensen@hbs.edu.

Jensen, Michael. 1983. Organization Theory and Methodology. Accounting Review 58(2): 319-339. Jensen, M., E. Fama, J. Long Jr., R. Ruback, G. W. Schwert, C. W. Smith Jr. And J. Warner. 1989.

Clinical Papers and Their Role in the Development of Financial Economics. Journal of Financial Economics 24(1): 3-6.

Kamath, Ravindra. 1997. Long-Term Financing Decisions: Views and Practices of Financial Managers of NYSE Firms. Financial Review 32(2) 331-356.

Klein, Benjamin, Robert Crawford and Armen Alchian. 1978. Vertical Integration, Appropriable Rents, and the Competitive Contracting Process. Journal of Law and Economics 21: 297-326.

Kraus, Alan and Robert Litzenberger. 1973. A State-Preference Model of Optimal Financial Leverage. Journal of Finance 28: 911-922.

Lintner, John. 1956. Distribution of Incomes of Corporations among Dividends, Retained Earnings, and Taxes. American Economic Review 46: 97-113.

Marcus, Alan. 1983. The Bank Capital Decision: A Time Series-Cross Section Analysis. Journal of Finance 38(4): 1217-1232.

Masulis, Ronald. 1988. The Debt/Equity Choice. Ballinger Publishing Company. Merton, Robert. 1974. On the Pricing of Corporate Debt: the Risk Structure of Interest Rates. Journal of

Finance 29(2): 449-470. Milgrom, Paul and John Roberts. 1992. Economics, Organization and Management. Englewood Cliffs

(NJ): Prentice-Hall. Miller, Merton. 1977. Debt and Taxes. Journal of Finance. 32(2): 261-275.

Miller, Merton. 1988. The Modigliani-Miller Propositions after Thirty Years. Journal of Economic Perspectives 2(4): 99-120.

Miller, Merton. 1995. Do the M&M Propositions Apply to Banks? Journal of Banking and Finance 19(3-4): 483-489.

Modigliani, Franco and Merton Miller. 1958. The Cost of Capital; Corporate Finance and the Theory of Investment. American Economic Review 48(3): 261-297.

Modigliani, Franco and Merton Miller. 1963. The Corporate Income Taxes and the Cost of Capital: A Correction. American Economic Review. 53(3): 433-443.

Myers, Stewart, and Nicholas Majluf. 1984. Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have. Journal of Financial Economics 13(2): 187-221,

Myers, Stewart. 1977. Determinants of Corporate Borrowing. Journal of Financial Economics 5(2): 147-175.

- 28 -

Myers, Stewart. 1984. The Capital Structure Puzzle. Journal of Finance 39(3): 575-592. Myers, Stewart. 1993. Still Searching for Optimal Capital Structure. Continental Bank Journal of Applied

Corporate Finance 6(1): 4-14. Myers, Stewart. 2001. Capital Structure. Journal of Economic Perspectives 15(2): 81-102. Opler, Tim and Sheridan Titman. 1996. The Debt-Equity Choice. Working Paper no. 97-9. Columbus:

Charles A. Dice Center for Research in Financial Economics, Ohio State University. Orgler, Yair and Robert Taggart. 1983. Implications of Corporate Capital Structure Theory for Banking

Institutions. Journal of Money, Credit, and Banking 15(2): 212-221. Osterberg, William, and James Thomson. 1996. Optimal Financial Structure and Bank Capital

Requirements: An Empirical Investigation. Journal of Financial Services Research 10(4): 315-332.

Petersen, Mitchell and Raghuram Rajan. 1994. The Benefits of Lending Relationships: Evidence from Small Business Data. Journal of Finance 49 (1): 3-37.

Rajan, Raghuram and Luigi Zingales. 1995. What Do We Know about Capital Structure? Some Evidence from International Data. Journal of Finance 50(5): 1421-1460.

Ross, Stephen. 1977. The Determination of Financial Structure: The Incentive-Signalling Approach. Bell Journal of Economics 8(1): 23-40.

Ross, Stephen. 1988. Comment on the Modigliani-Miller Propositions. Journal of Economic Perspectives 2(4): 127-133.

Saá-Requejo, Jesús. 1996. Financing Decisions: Lessons from the Spanish Experience. Financial Management 25(3): 44-56.

Santos, João C. 2001. The Regulation of Bank Capital: A Review of the Theoretical Literature. Financial Markets, Institutions and Instruments 10(2): 41-84.

Santos, Mario C. 2000. Firm’s Capital Structure Decisions: Theory and Empirical Evidence from Portuguese Banks. Unpublished Ph.D dissertation. Universidade de Aveiro, Portugal.

Scott, James. 1977. Bankruptcy, Secured Debt, and Optimal Capital Structure. Journal of Finance 32(1): 1-19.

Sharpe, Ian. 1995. Determinants of Capital Structure of Australian Trading Banks. Asia Pacific Journal of Management 12(2): 97-121.

Smith, Vernon. 1972. Default Risk, Scale, and the Homemade Leverage Theorem. American Economic Review 62(1/2): 66-76.

Stiglitz, Joseph. 1969. A Re-Examination of the Modigliani-Miller Theorem. American Economic Review 59(5): 784-793.

Stiglitz, Joseph. 1974. On the Irrelevance of Corporate Financial Policy. American Economic Review 64(6): 851-866.

Stiglitz, Joseph. 1988. Why Financial Structure Matters. Journal of Economic Perspectives 2(4): 121-126. Stiglitz, Joseph. 1989. Mutual Funds, Capital Structure, and Economic Efficiency: Discussion. In Theory

of Valuation - Frontiers of Modern Financial Theory, Volume 1, ed. S. Bhattacharya, G. Constantinides, 342-356. Savage: Rowman & Littlefield.

Stulz, René. 1990. Managerial Discretion and Optimal Financing Policies. Journal of Financial Economics 26(1): 3-27.

Tirole, Jean. 1999. Incomplete Contracts: Where Do We Stand? Econometrica 67(4): 741-781. Williamson, Oliver. 1990. The Firm as a Nexus of Treaties: an Introduction. In The Firm as a Nexus of

Treaties, ed. M. Aoki, B. Gustafsson, and O. E. Williamson, 1-25. London (UK): Sage Publications.

- 29 -

Table 1

Sample Demographic Characterization [number]

CEOs/bank 51 Banks 33 CEOs of ‘state-owned’ banks 15 CEOs of ‘privately-owned’ banks 36 CEOs of ‘listed’ banks 19 CEOs of non-Listed banks 32 CEOs of ‘de novo’ banks 24 CEOs of ‘established’ banks 27 CEOs of ‘underleveraged’ banks 23 CEOs of ‘overleveraged’ banks 28

Table 2 Banks Descriptive Statistics

[106 EUR] Average banks’ total assets 3.587 Standard deviation of banks’ total assets 4.482 Minimum banks’ total assets 61 Maximum banks’ total assets 23.212 Average capital ratio (book value) 0,085 Standard deviation of capital ratio 0,068 Minimum capital ratio 0,015 Maximum capital ratio 0,315

Table 3 Distribution of CEOs’ Tenure

Years Frequency Percent Cumulative Percent

2 8 15.7 15.7 3 13 25.5 41.2 4 7 13.7 54.9 5 8 15.7 70.6 6 9 17.6 88.2 7 3 5.9 94.1 8 1 2.0 96.1 9 0 3.9 100.0

10 2 100.0

- 30 -

Table 4

CEOs Objective Function

Mean Scores Achieve and maintain high debt ratings 3.6 Achieve a capital structure similar to that of other banks 2.5 Enhance and sustain financial flexibility, independence, & survival 4.3 Maximize the market price of bonds and stock 2.3 Maximize the market share (in terms of net total assets) 2.4 Maximize the Price Earnings Ratio 2.1 Maximize the Return on Investment 3.6 Maximize the Return on Equity 4.8 Maximize the growth of earnings per share 3.5 Maximize shareholders’ returns 4.3 Maximize the book value of a share of stock 2.8 Maximize the book value of the bank’s net total assets 2.2 Maximize the cash flow per share of stock 2.8 Minimize the risk of financial distress and bankruptcy 3.3 Minimize the bank’s cost of capital 4.0

- 31 -

Table 4a

CEOs Financial Management Objectives

‘State-Owned’ and ‘Privately-Owned’ Bank CEOs Mean Scores Maximize the market price of bonds and stock 1.3 2.8 Maximize the Price Earnings Ratio 1.5 2.4 Maximize shareholders’ returns 3.5 4.7 Maximize the book value of the bank’s net total assets 1.5 2.5 Minimize the risk of financial distress and bankruptcy 2.5 3.7 Minimize the bank’s cost of capital 3.1 4.4

‘Listed’ and ‘Unlisted’ Bank CEOs Mean Scores Achieve and maintain high debt ratings 4.6 3.1 Achieve a capital structure similar to that of other banks 3.0 2.1 Maximize the market price of bonds and stock 3.1 1.8 Maximize the Price Earnings Ratio 2.8 1.8 Maximize the Return on Equity 5.4 4.4 Maximize shareholders’ returns 4.8 4.0

‘Over-Capitalized’ & ‘Under-Capitalized’ Bank CEOs Mean Scores Achieve and maintain high debt ratings 3.1 4.2 Achieve a capital structure similar to that of other banks 2.1 2.9 Maximize the Return on Investment 3.2 4.1 Maximize stock book value 2.2 3.5 Maximize the book value of the bank’s net total assets 1.8 2.7 Maximize cash flow per share 2.3 3.5

‘De Novo’ and ‘Established’ Bank CEOs Mean Scores Achieve and maintain high debt ratings 2.8 4.3 Significant difference at the 5 percent level for one-sided tests.

Significant difference at the 1 percent level for one-sided tests. Significant difference at the 5 percent level for two-sided tests.

Significant difference at the 1 percent level for two-sided tests.

- 32 -

Table 5

Capital Structure Policy [unit: percent]

Follow a previously defined set of guidelines on financing policy 31.4 Achieve and adhere to a definite target for capital structure 5.9 Reach an optimal capital structure by comparison of economic costs and benefits 39.2 Keep to the financing pattern historically followed by the bank 2.0 Follow a pre-determined hierarchy in exhausting available strategic financing sources 21.6

[unit: percent] CEOs of: State-Owned

Banks Privately-

Owned Banks Follow a previously defined set of guidelines on financing policy 60.0 19.4 Reach an optimal capital structure by comparison of, both, economic costs and benefits

13.3

50.0

Significant difference at the 5 percent level for one-sided tests. Significant difference at the 1 percent level for one-sided tests.

Significant difference at the 5 percent level for two-sided tests. Significant difference at the 1 percent level for two-sided tests.

- 33 -

Table 6

External Determinants of Banks’ Capital Structure Decisions

Mean Scores Possibility of a takeover bid 1.8 Change in the ratio of public deficit to gross domestic product 1.5 Changes in the regulation and supervision framework 4.3 Private consumption behavior 1.6 Currency market behavior 1.5 Capital market performance 3.3 World economy performance 1.8 National economy performance 2.9 Political instability 2.3 Change in the dynamics of credit demand 3.1 Change in firms’ and investors’ income taxation 2.8 Legal restrictions on share repurchases 1.8 Interest rate changes 2.6

- 34 -

Table 6a

External Determinants of Banks’ Capital Structure Decisions

‘State-Owned’ and ‘Privately-Owned’ Bank CEOs Mean Scores Possibility of a takeover bid 1.0 2.1 Private consumption behavior 1.1 1.9 Currency market behavior 1.0 1.6 Capital market performance 2.2 3.8 World economy performance 1.4 2.0 National economy performance 2.3 3.2 Change in firms’ and investors’ income taxation 1.9 3.1 Legal restrictions on share repurchases 1.0 2.1 Interest rate changes 1.7 2.9 ‘Listed’ and ‘Unlisted’ Bank CEOs Mean Scores Currency market behavior 1.8 1.3 Capital market performance 4.1 2.8 World economy performance 2.2 1.6

‘De Novo’ and ‘Established’ Bank CEOs Mean Scores Interest rate changes 3.2 2.0 Significant difference at the 5 percent level for one-sided tests.

Significant difference at the 1 percent level for one-sided tests. Significant difference at the 5 percent level for two-sided tests.

Significant difference at the 1 percent level for two-sided tests.

- 35 -

Table 7

Internal Determinants of Banks’ Capital Structure Decisions

Mean Scores Get the bank’s shares listed 3.3 Ownership structure and managerial control 4.6 Tax economies related to factors other than debt financing 2.4 Size of free cash flow 2.1 Earnings per share (avoid earnings dilution) 2.6 Historical performance of bank’s shares 2.2 Dividend policy 3.1 Investment policy / Growth opportunities 4.0 Financing viability of strategic objectives 3.7 Assets’ risk 3.2 Tax economies associated with debt financing 3.0 Correct mispricing in past security issues 2.0 Issuing costs 1.9 Bank size 3.3 Avoid mispricing in future security issues 1.8 Covenants in debt financing contracts 1.6 Rates of taxation on investors’ income 1.9 Risk and costs of financial distress and insolvency 2.2 Managerial expectations for bank’s future performance 3.4 Restructuring of bank’s asset portfolio 2.6 Bank’s reputation 3.8 Changes in bank’s level of profitability 2.8

- 36 -

Table 7a

Internal Determinants of Banks’ Capital Structure Decisions

‘State-Owned’ and ‘Privately-Owned’ Bank CEOs Mean Scores Historical performance of bank’s shares 1.1 2.6 Correct mispricing in past security issues 1.4 2.3 Issuing costs 1.1 2.2 Avoid mispricing in future security issues 1.1 2.1 Covenants in debt financing contracts 1.0 1.8 Restructuring of bank’s asset portfolio 1.9 2.9

‘Listed’ and ‘Unlisted’ Bank CEOs Mean Scores Get the bank’s shares listed 4,6 2,4 Tax economies related to non-debt financing factors 3,0 2,0 Size of free cash flow 2,6 1,8 Earnings per share (avoid earnings dilution) 3,4 2,2 Historical performance of bank’s shares 2,6 1,9 Tax economies associated with debt financing 3,7 2,5 Correct mispricing in past security issues 2,6 1,7 Covenants in debt financing contracts 2,2 1,2 Rates of taxation on investors’ income 2,5 1,5 Bank’s reputation 4,6 3,4 Changes in bank’s level of profitability 3,8 2,3

‘Over-Capitalized’ & ‘Under-Capitalized’ Bank CEOs Mean Scores Size of free cash flow 1.6 2.7 Financing viability of strategic objectives 3.2 4.4 Assets’ risk 2.7 3.9 Tax economies associated with debt financing 2.6 3.3 Correct mispricing in past security issues 1.5 2.7 Bank size 2.8 3.9 Avoid mispricing in future security issues 1.4 2.4 Covenants in debt financing contracts 1.2 2.0 Restructuring of bank’s asset portfolio 2.2 3.2 Bank’s reputation 3.4 4.3 Changes in bank’s level of profitability 2.2 3.6

‘De Novo’ and ‘Established’ Bank CEOs Mean Scores Assets’ risk 2.8 3.6 Bank’s reputation 3.5 4.2 Significant difference at the 5 percent level for one-sided tests.

Significant difference at the 1 percent level for one-sided tests. Significant difference at the 5 percent level for two-sided tests.

Significant difference at the 1 percent level for two-sided tests.

- 37 -

Table 8

Synthesis of Statistically Significant Findings ARGUMENT FINDINGS

Capital Structure Policy Bank Financial Management Objectives Maximization of shareholder value. Enhancing and sustaining

financial flexibility. Minimization of a bank’s cost of capital. Adhering to an optimal capital structure.

Market timing 54.9% of CEOs admit bank’s stock price performance is relevant for timing new security offerings

Capital Structure Determinants External factors affecting capital structure decisions

Relevance of regulatory discipline. Irrelevance of market for corporate control disciplinary.

Internal factors affecting capital structure decisions

More concerned of ‘established’ banks’ CEOs with bank reputation. Asset risk is less important ‘de novo’ banks’ CEOs.

Taxation Tax Hypothesis

Tax considerations are not perceived as playing a significant role in banks’ capital structure decisions. Taxation at the bank level more important than at investors’ level.

DeAngelo and Masulis (1980) Hypothesis Tax-loss carryforwards were deemed as relevant for capital structure. Banks carrying tax-loss carryforwards are more likely to issue equity.

Insolvency & Bankruptcy Impact of financial distress and bankruptcy Bankruptcy proceedings in the banking industry consume more

time and involve more costs than in similar procedures for non-banking firms.

Too big to fail hypothesis Strong evidence consistent with the too big too fail proposition. Agency and Governance Regulatory discipline of capital adequacy Moderate preference for regulatory capital discipline Market discipline hypothesis Preference for market value-based ratios. CEOs of ‘listed’ banks

are more concerned with minimizing the cost of capital and the risk of financial distress than their counterparts.

Disciplinary role of debt CEOs of under capitalized banks are more concerned with the disciplinary role of debt as shown by the importance given to ratings.

Share repurchases and capital structure decisions Limitations on share repurchases under current Portuguese law, were not considered relevant for bank’s capital structure decisions

Motivations for issuing convertible securities Agency problems between shareholders and managers, and existing and future shareholders, were cited as the determinant factor, followed by lower financing costs.

Asymmetrical Information Signaling theory Results are consistent with the signaling theory that external

equity financing signaling effects of Leland and Pyle. Undervaluation hypothesis Stock price reactions to announcements of (voluntary) equity

issues and debt offerings are interpreted as negligible. Signaling with Debt Issuance A large majority of CEOs also indicated that they were unlikely to

engage in this signaling activity Private placements, covenants and information Agency costs of debt less severe that for non-banking firms. Security private placements and capital structure decisions

High degree of concordance in undertaking private placements of debt issues. Lower concordance with private placements of equity.

Security private placements and private information sharing

Costs associated with asymmetric information problems in banks’ new security offerings might not be as severe as usually hypothesized for non-banking firms.

Growth opportunities influence on capital structure

Growth opportunities influence capital structure policy. Moderate support for the underinvestment hypothesis.

Product & Factors Markets Reputation hypothesis Significant difference in the importance assigned to bank

reputation by CEOs of ‘the novo’ and ‘established’ banks.