2.2. THE MEANING OF THE TERM 'FINANCIAL SYSTEM' AND THE IMPORTANCE OF FINANCIAL SYSTEMS

2.2.1. Defining Terms and Topics: 'Financial Sector' Versus 'Financial System'

What is a financial system, and what does the term 'financial system' properly refer to? The answer to these questions is not obvious, as one can easily see if one imagines the situation of a young World Bank economist who is instructed by her superiors to fly to some faraway country and come back with an overview of the 'financial system' of that country. What would she look at and what should she look at? What would those who sent her on the mission be interested in knowing?

She would certainly come back with information concerning the central bank and the monetary system of the country, its banks, and its stock market, if there is one. But this information may not be all that her superiors want to know. A policy-relevant analysis should put first things first. At least at a conceptual level, the analysis of any financial system should start with people and their business endeavours; it should examine their financial needs and their demand for financial services, and it should identify the ways in which these needs are satisfied and their demand is met. Accordingly, the underlying questions are:

1.	How do people accumulate wealth and transfer income over time? Which options are available and how are they used?
2.	How do businesses obtain financing if they, or their owners or managers, have attractive investment opportunities for which their own funds would not be sufficient?
3.	How do people cope with risks?

Addressing these questions raises others, some of which may be more familiar: Are there institutions in that country which meet the demand for savings opportunities, credit facilities, and risk management or insurance services? And if they exist, what is the nature of those institutions, how do they function, who owns and governs them, how do they operate, and how are they designed and regulated? And most importantly, what role do those institutions play in helping real people to solve their financial problems? How important is external financing in comparison to internal financing? How important are real savings or real investment, as means of storing and transferring wealth over time, in comparison to financial savings? And how important is formal insurance relative to other methods of risk mitigation, that is, those which involve measures confined solely to the real sectors of the economy? How important are formal, that is, regulated and supervised financial institutions and organized markets in comparison to informal financial relationships, such as those that may exist between friends and family members? Are the financial institutions and the organized markets 'good' according to some suitable standard? And is the legal and regulatory system conducive to the provision of financial services and the establishment of financial institutions and their proper functioning?

These questions are equally important for all countries, irrespective of whether one looks at Indonesia, Poland, or Germany. As the young World Bank economist will hopefully have come to realize after a few days, the formal financial institutions may not be the most important part of the financial system of any of these countries. These considerations motivate our distinction between two concepts and their definitions.

1.	The narrow concept is that of the financial sector. We define the financial sector as that part—or sector—of an economy which offers and provides financial services to the other sectors of the economy. It consists of the central bank, other banks, non-bank financial institutions, organized financial markets, and the relevant regulatory and supervisory institutions.
2.	The broader concept is that of the financial system. It can be defined in general terms as the interaction between the supply of and the demand for the provision of capital and other finance-related services.

In addition to the supply side, that is, the financial sector and its activities, the concept of the financial system also includes the demand side, and thus the economic units which may demand financial services. The demand for these services comes from many economic units, but most importantly from households which accumulate wealth or simply carry income over from one time period to the next, and from firms which need capital for investing.

The widening of the focus is more than a change of definition, since the concept of the financial system also includes units in the non-financial sector in so far as they do not demand, or do not succeed in obtaining, services from the formal financial sector. Indeed, if we wish to understand the financial system of a country, we must also indicate the extent to which, and the forms in which, households and other surplus units undertake real investment merely in order to accumulate and transfer assets; how firms and other deficit units obtain financing directly from households and other surplus units or engage in self-financing; and how households and firms protect themselves against risks. The state is also a part of the financial system, not only because it too supplies and demands financial services, but also because it serves as the organizer and regulator of the financial sector.

There is a second dimension in which the concept of the financial system is broader than that of the financial sector. Financial flows are mirrored in flows of information and flows of potential and actual influence. From the perspective of new institutional economics, there are obvious reasons why this is so, and it is also obvious that financial flows, information flows, and flows of influence are interdependent: Each of the three types of relationship plays a key role in determining the nature of the other two types. Actual and potential flows of information and influence constitute the essence of 'the corporate governance system'. As finance without governance would scarcely be possible, the corporate governance system is therefore an integral part of the financial system of any country.

2.2.2. The Importance of Having a Good Financial System

For a long time it has been common knowledge among economists that a well-functioning financial system is important for welfare and economic growth. Hilferding, for example, emphasized the role of the German banks in supporting the belated growth of industry in Germany. Schumpeter's view of the banker as the partner and supporter of the entrepreneur is well known, as is Gerschenkron's historical account of this role. Practitioners in banking and business as well as politicians who are active in this area will typically share the view that 'finance matters'.

The conventional view of the financial system as a factor contributing to growth lost its prominence in academic circles at the time when the neoclassical theory of economic growth gained general acceptance, as the analytical structure of this theory does not accommodate banks or the financial sector in general. Moreover, while some economists continued to consider financial development as an engine of growth, others claimed that it was not a cause but rather a consequence and a symptom of real economic development. For example, Hicks (1969) argued that the financial system of the United Kingdom played an important role in the Industrial Revolution, while Robinson (1952) questioned this view and suggested that the financial system developed as a result of economic growth, that is, that the causality went in the other direction.

The discussion about the role and importance of financial systems was revived around 1990. The World Development Report of 1989, which was specifically dedicated to this topic, provided the first econometric evidence that finance matters for growth and development; and it also suggested a causal link running from finance to growth. This strand of research culminated in the work of King and Levine. As the title of one of their influential papers indicates, they claim that 'Schumpeter might be right' (King and Levine 1993). More recent work, which is summarized in Levine (1997) and Winkler (2001), points in the same direction.

If one looks at this literature in detail one can easily recognize that, in its search for academic rigour, it does not, in fact, refer to the entire financial system in the broad sense introduced above, but more narrowly to the financial sector. However, the finding can easily be generalized using arguments such as those put forward in the work on 'law and finance' by La Porta et al. Thus, at least at the current state of the debate, one can safely conclude that a good financial system seems to be conducive to economic welfare and growth and that, therefore, the topic of this book is indeed an important one. But this then raises the question of standards of assessment. We will address this question in the following section.

2.3. DIFFERENT APPROACHES TO ANALYSING FINANCIAL SYSTEMS

As there are many contexts in which the financial system is considered to be relevant, it is not surprising that there are also many different ways in which financial systems are described and analysed in the literature. Each of them reflects not only a different focus, but also a different concept of what is, and what is not, part of the financial system. However, there does not seem to exist a common view of how one should describe financial systems.

In what follows, we shall briefly characterize four ways of describing and analysing financial systems. Let us refer to them as the 'institutional approach', the 'intermediation approach', the 'functional approach', and the 'systemic approach'. As will hopefully become clear, the four approaches are not mutually exclusive. Instead, they build on each other at least to a certain

extent. We will conclude this section by indicating why we advocate using the most comprehensive approach, and by demonstrating in which sense it is in line with the broad concept of the financial system introduced above.

2.3.1. The Institutional Approach

As the term suggests, the institutional approach focuses on the existing institutions such as the banks, insurance companies, investment and pension funds, financial markets, and the central bank system. Descriptions and classifications of those institutions which in their totality make up the financial sector of a country are the dominant topic in the literature which follows this approach. Thus, the institutional approach is narrow in its focus, and it is not analytical. This characterization of the institutional approach is not intended to be demeaning, since knowledge of the facts is a prerequisite for any attempt to dig deeper.

In addition to descriptions, classifications and statistical material, work which follows this approach can, and typically does, also offer comments and assessments pertaining to certain features of the various institutions within the financial sector or of the sector as a whole. For instance, it might be supplemented by an evaluation of the degree of competition in the banking sector, of its openness to foreign competitors and of the competition within and between different groups of financial intermediaries; or the transparency of price determination and investor protection in financial markets; or the quality of regulation and supervision.

2.3.2. The Intermediation Approach

Although, strictly speaking, no approach can be 'purely descriptive' in so far as any description involves the application of at least some analytical criteria which determine what to select and report, it is nonetheless true to say that the institutional approach is primarily descriptive in the sense that it is not explicitly shaped by theoretical considerations concerning, for example, the economic functions of the financial sector and its constituent institutions and their role in the national economy. Conventional economic approaches are different in this respect. One of these, which we do not wish to discuss at this point, is the monetary approach. It regards the financial sector, which is composed of the central bank and the commercial banking system, as the apparatus which supplies money to the real economy.

There is another conventional economic approach which can be traced back to the work of Gurley and Shaw and which emphasizes a different role of the financial sector, namely that of intermediation and transformation. In its original form, this approach, which we call the 'intermediation approach' for convenience, views the financial sector as composed of banks and other financial intermediaries. Intermediaries go between 'surplus spending units', which have more income in a given time period than they themselves want to spend on consumption and real investments during that period, and 'deficit spending units', which want to spend more than they have in terms of current income and income carried over from earlier time periods. In simple terms, the surplus units are savers, which is the characteristic role of households, and the deficit units are investors, the typical role of firms, and in many cases also of the state. Intermediaries substitute for direct finance by accepting some form of deposits from the savers and providing loans or equity to investors.

Intermediation helps to alleviate some of the problems which would make direct finance difficult. The main problem is that of divergent preferences and economic needs. Typically, savers prefer to have their funds available at short notice, to invest small amounts for relatively short terms, and to have their funds used for purposes which involve a low level of risk. Typical investors want exactly the opposite: They need funds without the threat of having them withdrawn unexpectedly; they want to have larger amounts for a longer period of time; and they want to be able to invest them in assets which may include some that entail a higher level of risk. In addition, there is a problem of information and incentives, or of adverse selection and moral hazard, which makes direct finance difficult. Being less informed and less capable of monitoring the use of their funds, savers might simply abstain from directly financing investors, or request compensations which make external financing unattractive for investors.

Intermediaries perform the following functions: They can provide liquidity; they can compensate for a certain mismatch of maturities and amounts, that is, perform what is referred to as term-to-maturity transformation and unit-size transformation; they can act as experts in assessing the economic potential of those who seek funding; and finally they serve as 'delegated monitors' (Diamond 1984, and the extensions in Diamond and Rajan 2001, and Tyrell 2003). In performing these functions, the financial sector can create value for the economy.

In an extended version, this approach also takes account of the economic role of financial markets and in particular secondary markets, and thus of the securitization of financial assets. Financial markets can also perform some of the transformation functions listed above. By doing so, they also mitigate the tensions which would stand in the way of transferring financial surplus from those who want to save to those who have good investment opportunities. Moreover, intermediation in the strict sense of on-balance-sheet intermediation can be combined with intermediation via markets in various ways. Allen and Santomero (2001), for example, show for the case of the United States how the role of banks has shifted from acting as financial intermediaries themselves to facilitating access to capital markets and to the transformation services that the capital markets provide to the economic units of the non-financial sectors of the economy.

This concept of what financial intermediaries and financial markets do suggests a way of describing and analysing a given financial system in terms of the extent and the effects of intermediation, transformation, and securitization. Evidently, such an approach is much more theory-based than the institutional approach. It not only covers the financial sector, but also includes, as an essential element, the non-financial sectors. Thus, it deals with the financial system as a whole. Moreover, it lends itself relatively well to measurement (see Hackethal and Schmidt 2000, 2003) and suggests a straightforward standard for evaluation: A financial system is 'better' if it provides more intermediation in the narrow and in the broad sense. However, the intermediation approach is quite selective in terms of what it takes into consideration, which indicates that one could usefully combine it with further extensions. The two approaches to which we turn now can be regarded as relevant extensions.

2.3.3. The Functional Approach

While the institutional approach is useful in its emphasis on the given institutional structure, it goes too far in this respect and has some serious weaknesses. As Merton argues, it cannot explain how and why the institutional structure of a financial system changes and how it could evolve over time. More generally, it lacks a deeper understanding of the functions a financial system should perform. While the intermediation approach does have a firm theoretical basis and emphasizes functions, it is not entirely satisfactory in so far as it only focuses on two functions of the financial sector, namely intermediation and transformation.

Robert C. Merton and various co-authors have developed an alternative to the institutional perspective and, one might add, a generalization of the intermediation approach. What they call the functional approach focuses on functions rather than institutions as given and, therefore, as the conceptual point of departure. In this approach five functions are identified and distinguished from the primary or core function at a less abstract and aggregate level. A financial system provides

(1)	a way to transfer economic resources through time and across regions and industries, which is singled out as the core function of any financial system;

(2)	a payments system to facilitate the trade of goods, services, and financial claims;
(3)	a mechanism for pooling resources and for subdividing shares in large-scale indivisible enterprises;
(4)	a way to manage uncertainty and control risk;
(5)	price information that can be useful in coordinating decentralized decision-making;
(6)	a way to deal with asymmetric information and incentive problems.

As its advocates claim, the functional approach offers a useful frame of reference for analysing a country's entire financial system; it is more reliable and generates more insights than an institutional approach because the basic functions of a financial system are essentially the same in all economies and are more stable than the identity and structure of the institutions performing them. According to this approach, the analysis of a financial system starts by describing the functions performed by the different elements of a country's financial system and determining how the functions are currently being performed (see Merton and Bodie 1995: 17).

The functional approach has the virtue of abstracting from institutional detail and instead focusing on the underlying economics. However, it may also go too far. A strict application of this approach presupposes that functions are separable, and this assumption may be wrong: On the one side, an arrangement in which a particular type of institution performs multiple financial functions at the same time may be efficient, and on the other side, specific configurations of institutions may be required to reduce frictions which would otherwise hamper the allocation of resources, and thus lower the performance of a financial system. This requirement, namely the complementarity between different elements (institutions) of a financial system, will be discussed in more detail in the next subsection. As a result, there are major trade-offs and interrelationships between the different functions described above and the different institutions performing those functions (see Allen and Gale 2000 for a thorough analysis of these trade-offs). In this respect different financial systems may have different strengths and weaknesses, which the functional perspective seems hardly capable of grasping.

2.3.4. The Systemic Approach

A systemic approach aspires to accomplish just that: It describes and analyses a financial system in terms of the interrelations between its elements and the impact which these interrelations have on the performance of the system as a whole. In a systemic perspective, a financial system is an ordered set of complementary and possibly consistent elements or subsystems.

What is the meaning of 'complementarity'? Two (or more) elements of a system are complementary to each other

(1)	if the positive effects of the values taken on by the elements mutually reinforce each other in terms of a relevant evaluation function, and the negative effects mutually mitigate each other; that is
(2)	if a higher value for one element increases the benefit yielded by an isolated small increase in the value for the other element (and vice versa); and
(3)	if, as a result, the 'quality' or the 'economic value' of a system depends on the values taken on by its (complementary) features being consistent with each other or, to put it simply, on the values 'fitting together'.

Whether the elements or subsystems of a given system are complementary is determined by theoretical considerations. In the case of financial systems, such considerations would have to show that elements such as the role of banks in the financial sector, the financing patterns of corporations and the governance structure of big corporations, as well as the role of financial markets as compared to that of intermediaries, the structure of the pension system and certain parts of the legal system are complementary to each other. This can indeed be shown to be the case. Thus, it is accurate to say that all financial systems are shaped by the complementarity of their main elements. Moreover, theoretical considerations allow the conclusion that there may not be more than a few or even only two 'efficient' types of financial systems, as is reflected in the well-known classification of financial systems into two groups, depending on whether they are perceived as being bank-based or capital market-based.

However, complementarity implies merely that a potential exists: Economic benefits may accrue, but only if the main elements of a (financial) system take on values which fit well together. This potential will not necessarily be realized. This is precisely the aspect covered by the concept of 'consistency'. We refer to a system made up of complementary elements as being consistent if the benefits of complementarity are exploited and if, therefore, a small change in the value taken on by an individual feature, or by several features, does not permit an improvement in terms of the objective function or the evaluation function. Whether or not a given financial system is consistent is an empirical question.

As a consequence, the financial system of a country can be characterized by determining which of the known types of financial systems (defined on the basis of their complementary elements) it conforms to—if any at all—and by investigating to what extent its elements are indeed consistent.

Of course, these definitions of complementarity and consistency presuppose that there is a standard of measurement with which systems can be evaluated—at least approximately—in terms of the benefits they offer, and that the values taken on by the elements can be measured at least in the weak form of an ordinal ranking and/or that it is possible to distinguish between 'polar' values. The standard for assessment might be a general level of economic welfare or the contribution which the financial system makes to economic growth; but it might also be the extent to which the functions of financial systems listed by Merton and his co-authors are performed, and also how well they are performed.

A consistent system of complementary elements is in a state of (static) equilibrium. Typically, the key elements of systems characterized by complementarity include multiple equilibria or local optima. In most cases it is impossible for the individual actors operating in the system to judge whether a local optimum is also the global optimum. This feature has important implications for the development of financial systems and especially for the issue of the convergence of different financial systems over time. With respect to the German financial system this aspect will be taken up in the final chapter of this book.

The twin concepts of complementarity and consistency would seem to give a more precise meaning to what is colloquially called 'a real system', as it is highly plausible to assume that one's assessment of individual elements, and indeed of a specific change in one element, depends in a crucial way on the values which the other elements of the system take on. What are the implications of this insight for the task of describing and analysing a given financial system? Unfortunately, a formal proof of complementarity and consistency based on the mathematical theory which underlies the theory of complementarity cannot be performed in practice because it would require much more information than is available. However, one can attempt to describe a given financial system informally in such a way that complementarities which are presumed to exist become visible. If one can show that different key elements of the financial system in question fit together in a specific way, then this system is also likely to be consistent, and it can be classified as belonging to one of the two types of financial system distinguished above. Thus, this investigation should be a part of the characterization of any financial system.

2.3.5. Implications

All four approaches discussed above have their merits, and each of them is taken into account in the following chapters of the book as well as in the following section of the present chapter. However, as the systemic approach is the most comprehensive one, we will now attempt to outline the most important features of the German financial system using the concept of complementarity. We will argue that these features together constitute, or at least did constitute until quite recently, a system of complementary elements, each of which fits together with the others, thus making the others more effective than they would otherwise be.