Frameworks for Monetary Stability

21 Adapting Central Banking to a Changing Environment

Carlo Cottarelli, and Tomás Baliño
Published Date:
December 1994
  • ShareShare
Show Summary Details

The debate about central banking, which is as old as central banks themselves, has been given fresh impetus by a number of recent developments. In Europe, an unprecedented project to set up a supranational central bank has become an integral part of the European Community treaties. The process of adapting the institutional arrangements of former socialist states to the needs of a market economy invariably has meant establishing a central bank along Western lines. New central bank statutes have just come into force in France, while in the United Kingdom a debate on the central bank’s present institutional mandate is under way. In the United States, as well as in other countries, the question of whether the central bank should be kept out of the supervisory business is being discussed. Academic journals, for their part, abound with analyses of central banks’ handling of monetary policy and assessments of the degree of central bank independence.

Attempts to define the essence of central banking in simple and immutable formulas inevitably leave out something crucial. In Walter Bagehot’s time its essence was considered to consist in lending of last resort; today, central bankers’ views of the desirability and perhaps of the very meaning of this function are considerably diversified. Those of us who studied economics in the 1960s and early 1970s were taught that monetary policy is the main, if not the only, concern of central banks. Today no one would deny that payment system concerns have been important all along, and that in many countries banking supervision has been no less important than monetary policy in shaping the functions and organization of the central bank.

The subject calls for pragmatism. Thus, rather than trying to fit some “ideal model” to a recalcitrant reality, it is better to take the route of asking what practical problems central banks were meant to address in the various stages of their history and why one solution, rather than another, prevailed in the given circumstances of the time. Such an approach reveals that flexibility, or adaptability to changing needs, is a key feature explaining the continuing existence and good health of this three-centuries-old social institution.

Today we are once more in the midst of great changes in the financial and economic environment, calling central banks to a new effort of adaptation. The reasons for this transformation and the direction this effort is taking are the subject of this paper. It first describes the evolutionary logic behind the emergence of present-day central banks, with the three basic functions they perform. It then examines the new challenges and why they have arisen. Finally, it offers some reflections on the current and prospective lines of adaptation. The paper has no ready-made ideal model to which a central bank should conform. Rather, it attempts to read in the present in order to get as comprehensive a view as possible of the sense and direction of the evolution under way.

The Functions of Central Banks

The Triadic Nature of Central Banking

In 1967 Sir John Hicks summarized his views on the working of a monetary economy in a well-known series of lectures, to which he gave the title “The Two Triads” (Hicks, 1967a). The gist of his argument was that money is inherently a complex phenomenon, in terms both of the three functions it performs—means of payment, unit of account, store of value—and of the three motives people have to hold it—for transaction, precautionary, or speculative purposes. What Hicks meant to underline is the strong economic and historical connection between the three elements of each triad, so that neglecting any one function of money or motive for holding it involves the risk of simplifying the analysis to the point of making it useless or even misleading.

Now, what I want to suggest is that central banking also has a triadic nature. The activity of modern central banks spans three areas: monetary policy, the payment system, and banking supervision. Moreover, a close relationship exists between these three areas and the three functions of money. Monetary policy seeks to stabilize the real value of the unit of account; banking supervision serves to protect savings invested in monetary assets, or money as a store of value; and payment system policy is intended to make money an efficient and safe means of payment. Finally, these diverse and apparently unrelated activities are in reality strongly interconnected, not only in an economic way but also in a deeper institutional sense because they share the art of reconciling monetary rigor and flexibility of response in the face of unexpected shocks.

Even though central banks have evolved over time, gradually broadening the scope of their activities, the social function they perform has remained basically the same, as Hicks himself pointed out in his classic reappraisal of Harry Thornton’s Paper Credit (Hicks, 1967b). The evolution of central banks then can be seen as a process of ongoing adaptation, driven by the need to ensure effective execution of their basic functions in the context of ever-changing financial markets and increasingly complex economies (Giannini, 1994). In this light, the challenge that confronts us today is to adapt central banks to keep pace with the enormous changes that have occurred in recent years in the financial, economic, and technological environment in which they operate.

Payment Technologies as the Driving Force

Modern central banks have their origin in a fundamental change in monetary systems: the transition from a payment technology based on commodity money to one based on fiduciary money. In a fiduciary payment technology, money is intrinsically useless. Payments in fiduciary money differ from those made with commodity money in that they are carried out by transferring liabilities of a third party. The value of these liabilities, and hence the quality of payments, depend on the standing of the third-party issuer. This is why a system based on fiat money requires an ultimate issuer whose liabilities are universally accepted and trusted. It also explains why, after some not entirely satisfactory experiences with unregulated private issuers, it was recognized that the role of ultimate issuer should be formally entrusted to a “central” bank set up to perform a public function. Thus, central banks were created—though not strictly out of a conscious design—to issue liabilities that would be accepted by everyone in the economy and, by their existence, to underpin confidence in a payment system based on fiduciary money and ensure its integrity and efficiency.

The evolution of central banks went through two major phases. The first, associated with the spread of the convertible bank note, led to the recognition that the issue of notes had important social implications and therefore required protection of the public interest through state-enforced restrictions on issuing banks, which in most countries nonetheless remained private concerns aiming at profit-maximization. The second phase, associated with the rise of bank deposits, led to the suppression of the profit motive and the recognition that protecting the stability of the banking system required a public institution endowed with lending-of-last-resort powers.

As a result of the spread of commercial bank money, however, central banks have come to play another essential, though less visible, function: the supply of payment services to banks similar to those that banks provide to individuals and firms. When payments occur between banks (and it should be noted that any payment between two parties not holding accounts with the same bank also involves an interbank payment), the central bank is the only possible third party—in the sense that it is external to the banking system as a whole. The transfer of liabilities of the central bank is the only payment that instantaneously and totally extinguishes an obligation and closes the transaction, thus eliminating all risk. Central bank money held by banks, or bank reserves, is excluded from the overall money stock, so that it is scarcely visible, but it nonetheless acts as the indispensable lubricant of the entire monetary system (Angelini and Passacantando, 1992). As a result, present-day payment systems can be seen as a pyramid, with individuals and firms forming the base; then commercial banks providing them with payment services; and finally, at the apex, is the central bank, which plays the same role vis-à-vis commercial banks that the latter play vis-à-vis individuals and firms.

Managed Money and the Rise of Monetary Policy

By the end of the 1920s this structure of the payment system and the role played in it by the central bank were firmly established throughout the industrialized world. In the 1930s, under the pressure of the Great Depression, currents of thought that had developed earlier came to the fore; these emphasized the role of the central bank in the determination of the overall level of economic activity. Monetary policy came to be seen as part of a broader array of economic policy instruments. After decades, even centuries, of relatively passive attitude, managed money came of age. By the end of World War II, the idea that governments had a responsibility for macroeconomic stabilization and that monetary policy ought to be used for this purpose had gained wide acceptance.

For a time, convertibility remained a key feature of the monetary system: the amount of fiat money was, or was supposed to be, proportional to the amount of gold. In the long run, however, the notion of managed money—or “monetary nationalism,” as Hayek (1939) preferred to call it—proved incompatible with the notion of an objective anchor. In truth, convertibility was placed at the heart of the Bretton Woods system. But, as Guido Carli once said, it was little more than a fiction, which was “stretched beyond credibility” and finally abandoned at the beginning of the 1970s under the pressure of the growing balance of payments deficits of the reserve-currency country.

The advent of fiat money led to a new policy responsibility—that of managing the money supply for macroeconomic purposes. In line with this aim, between the late 1930s and the 1950s most central banks were nationalized or declared public institutions. The new responsibility required the elaboration of a conceptual model to guide the daily activity of central banks. Two analytical tools were of crucial importance in this respect: Hicks’ portfolio approach to money demand, which stressed the role of money as a store of value (Hicks, 1935), and the theory of the money-multiplier, which clarified the link between central banks’ liabilities and the overall supply of money. On the operational side, traditional instruments, such as the discount window, were gradually superseded by open market operations.

The growing importance of monetary policy and the development of an analytical approach that played down the means-of-payment and unit-of-account functions of money gradually de-emphasized the central bank’s payment system role, though this continued to take up a large proportion of central bank personnel and infrastructure.

The Role of Banks and Banking Supervision

Just as monetary policy is linked to the decline of convertibility, so banking supervision is linked to the rise of bank deposits in the total money stock. Confidence in the perfect substitutability between commercial and central bank money is a precondition for the former becoming a widely accepted medium. And since the burden of ensuring that substitutability rests ultimately with the central bank, the central bank needs to verify and promote the good health of commercial banks.

The link between banking supervision and the other two elements of the triad of central bank functions is equally clear. With bank deposits representing the bulk of the money supply, the banking system became not only the key actor in the payment system, but also the main channel of transmission of monetary policy impulses to the real economy. Being in a position to gauge the soundness and composition of banks’ balance sheets became a prerequisite for an effective monetary policy. The wave of bank failures that swept the United States in the 1930s, owing in part to the unduly restrictive stance of monetary policies, is a textbook example of the potentially disastrous consequences of neglecting the state of health of the banking system.

Until the opening decades of this century there was little appreciation of the need to endow central banks with supervisory powers over commercial banks. The first central bank to be formally entrusted with this function was probably the Federal Reserve, in 1913, although it refrained from using its supervisory powers for years, at least until the banking reform of the 1930s. Other central banks began supervising banks in the 1930s, either directly or through so-called Banking Inspectorates. The Bank of England was formally assigned responsibility for bank supervision only at the end of the 1970s, and it is well known that in some countries, such as Germany, the central bank does not exercise supervisory powers, although it constantly monitors the state of health of individual banks. These exceptions need to be evaluated in the light of the specific national financial structure and the de facto situation.

The case of the United Kingdom may be cited in this regard. The Bank of England assumed a public status at an early date compared with other central banks—not later than the 1880s. At the time, however, no need was felt to underscore this new status by formalizing the powers of the central bank or changing its ownership structure. Things happened quite naturally. Fred Hirsch explained very clearly how the oligopolistic structure of the London financial market, the club-like attitude of the ruling financial elite, and the reputation the Bank had acquired over time made acceptance of the Bank of England’s leadership uncontroversial (Hirsch, 1977). The Bank’s officials never lacked up-to-date information and the effectiveness of their actions never suffered because of the informal nature of their powers. This state of affairs lasted until the 1970s, when increasing competition in British financial markets and the growing presence of foreign-based institutions cast serious doubts on the effectiveness of informal controls. The subsequent formalization of supervisory powers, as a consequence, marked a turning point in the British approach to central banking.

A new debate on the appropriateness of attributing banking supervision to the central bank has developed recently in France, the United Kingdom, and the United States. Going deeply into this matter would exceed the limits of this paper: suffice it to recall that conflicts of interest and moral hazard are cited as motives for breaking the triad. No doubt it is technically feasible to take supervisory responsibility away from central banks, just as it is conceivable to separate the unit of account from the means of payment, to return to the analogy I made between central banking and money. But is it desirable to do so? In my view, the arguments that have been put forward so far do not carry decisive weight. A conflict between price stability and bank stability hardly arises; and, if it does, a careful weighing of the pros and cons of alternative policy courses is more likely to come from a single institution than from coordination between separate agencies, which is always difficult. Similarly, a degree of moral hazard is probably unavoidable in certain policy functions. On the other hand, intimate knowledge of the state of banks is indispensable even for a central bank not formally entrusted with the banking supervisory function. And the effectiveness of this function can be greatly enhanced if the supervisor is the central bank—i.e., the institution that has a direct knowledge of how banks, the money market, and the payment process operate.

The New Challenges

The triadic configuration presented so far remained more or less stable for about 50 years, since the 1930s, when the implications of its underlying causes—the transition from commodity to fiat money and the rise of bank deposits—unfolded in full.

Today, we are again in the middle of far-reaching changes, engendered primarily by two interrelated forces: technological change (in the twofold sense of financial innovation and data processing and communications) and the loosening of the coextensiveness between economic and political frontiers. These changes are altering the way payment systems and financial markets work; at the same time, they are straining the governance structure of which traditional central banks are part, since they are eroding the very notion of “national” sovereignty as far as the economy is concerned. Let me briefly describe the main evolutionary trends set in motion by these forces.

The Abuse of Nonconvertibility

Fiat money carries greater risk of abuse than metallic or convertible money, since its supply is not constrained by an objective anchor. This risk is far from merely theoretical. After surveying the early experiments with irredeemable money, Irving Fisher (1920) concluded that it had “almost invariably proved a curse to the country employing it.” Indeed, the experiments Fisher had in mind were undertaken either by authoritarian governments hungry for funds or in the midst of a major national emergency, such as a war. The democratization of politics in the course of this century and the challenges posed by the Great Depression subsequently made these experiments appear less relevant.

Yet Fisher’s warning is not to be taken lightly, as we have learned through the painful experience of the inflationary outburst of the past decades. In another paper published in this volume (Chapter 2), Robert Flood and Michael Mussa show that there is a clear structural break in the behavior of the price level before and after the repeal of convertibility. While under convertibility the price level—albeit in some cases fluctuating widely—showed a tendency to return to its long-run level, in later decades it began to rise steadily. Nowadays the very notion of a stable price level seems devoid of practical content, even in the most “virtuous” countries. The most one can hope for is stabilization at a very low rate of inflation.

This somewhat dismal record cannot be blamed entirely, if at all, on monetary mismanagement, but it reflects deeper phenomena: the enormous increase in the economic role of government, the pressures of unionized labor, and the change in intellectual climate referred to above. As early as the 1930s, Keynes pointed out the desirability of stabilization policies. In later decades the role of government expanded well beyond stabilization, encompassing redistributive and allocative purposes. Government expenditure reached unprecedented peaks, as a percentage of gross national product (GNP). Although there is no logical connection between public expenditure and monetary expansion, the practical impact on monetary management of such a rapid expansion of the economic role of government is undeniable. The subjugation of monetary policy to the needs of interventionist economic policy was so complete that Jacques Rueff (De Gaulle’s influential monetary advisor) once lamented that “Monetary policy is no longer à la mode. Until recently it was even completely forgotten. The specialists who talked about money were considered retarded.”

In this intellectual climate, when the objectives of short-run stabilization and price stability came into conflict under the pressure of the first oil shock and unions’ rising demands, in the early 1970s, it is not surprising that price stability had to give way. Under the institutional arrangements in force at the time, central banks could do little to stem the tide.

The notable exception to the general trend was Germany, where the economic behavior of business, government and labor, as well as public opinion, consistently gave high priority to the preservation of price stability, even at times of high unemployment. In this environment, the central bank availed itself of its ample operational autonomy to give absolute priority to price stability, meaning a rate of inflation as close to zero as possible, in all circumstances. The result was that in the 40 years between 1952 and 1993 annual inflation in Germany averaged less than 3 percent and exceeded 6 percent only twice. The German experience—whose relevance has in the meantime been enhanced by the concurrent development of a theoretical literature linking the autonomy of the central bank to the credibility of anti-inflationary policies—suggests, on the one hand, that fiat money need not necessarily be “a curse” and, on the other, that institutional safeguards have to be strengthened if the objective of price stability is to be attained.

The Changing Financial Structure

In the spectacular transformation of the financial structure that we have witnessed in the last two decades, three main components can be identified: the disproportionately rapid development of financial markets and transactions compared with the real economy; the blurring of the frontiers of financial specialization; and internationalization. Leaving the latter aside for the moment, let me briefly discuss the first two components.

A few often-quoted figures are sufficient to give an idea of how much the financial system has outpaced the growth of the real economy. Between the late 1970s and the early 1990s, Goldsmith’s Financial Interrelations Ratio, which measures the size of the financial sector in proportion to real wealth, rose by more than 30 percent on average in the six most industrialized countries; in the United Kingdom and the United States, in particular, it rose by about 90 and 40 percent, respectively. In the same period, stock exchange capitalization increased, as a percentage of GNP, by approximately four times in France and Italy, by three in Japan and the United Kingdom, and by two in Germany; meanwhile, in these five countries annual shares issues have increased, on average, from about 0.8 to 1.7 percent of GNP.

The underlying causes of this evolution can be summarized as follows. First, the growth in the real economy has required an ever-increasing amount of financial transactions, as the specialization in production and distribution and the dissociation between centers of savings and investment have multiplied the number of transfers of money and wealth associated with any unit of output. Second, within the financial sphere itself, a growing share of financial operations has gone through the market instead of being confined within the balance sheet of an institution, whether a bank or another intermediary.

The other phenomenon modifying the financial structure is the blurring of the boundaries between different financial contracts and institutions. The regulatory framework and the legislation inherited from the 1970s were based on a double tripartition: on the one hand, three basic contractual forms—equity, debt, and insurance; on the other, three types of financial institutions—banks, insurance companies, and capital markets. There was a fairly close correspondence between the two tripartitions, although the match was never perfect since, for example, capital markets also dealt in debt contracts and banks in many countries could deal in equity contracts.

Things have become far more complex in recent years, as a growing part of financial business, probably most of it by now, is carried out in mixed forms that hardly fit our conceptual boxes. This can be seen by looking at the recent evolution of contractual forms: the spread of complex contracts that combine features of debt, equity, and insurance, such as convertible bonds, indexed bonds, floating-rate notes, and so on; the unbundling of the constituent parts of a given contract, which consequently can be priced explicitly and traded separately; and the growth of off-balance-sheet items, (i.e., contracts that hedge the buyer against liquidity risk but do not give rise to a visible financial transaction at the time they are signed).

Similarly, in part as a result of deregulation in the 1980s, financial institutions have considerably expanded the scope of their activities, to the point that it may be less than obvious where a bank ends and a securities house or an insurance company or, for that matter, a mutual fund begins. Much attention has been devoted in recent years to examining whether these trends imply a convergence of the financial structure of the major industrial countries toward a single model. The enormous increase in the volume of trade in financial markets and the marked process of bank disintermediation in some countries apparently suggest an increasing orientation of the financial structure toward markets rather than institutions (Folkerts-Landau and Garber, 1992). However, it has also been noted that the trend toward securitization in the Anglo-Saxon world stemmed from the attempt of banks themselves to circumvent burdensome regulations by developing off-balance-sheet items; and that financial innovation has been far less pronounced in Germany and Japan, where banks were already allowed to undertake a wide range of activities (de Cecco, 1993).

Whatever the ultimate impulse that causes them, these developments imply a considerable erosion of one of the pillars of the traditional model: the strong tie between the central bank and the banking system. With the fading of operational boundaries between different classes of intermediary, all three of the functions performed by central banks are affected. Monetary policy ceases to hinge on the banking system as the sole or even primary transmission mechanism. Supervision aimed at stability is bound to grow less and less effective if its scope is not redefined to encompass nonbank financial institutions. Payment system policy must allow for the fact that the increasing negotiability of securities means that intermediaries other than banks are able and willing to offer payment services.

New Technologies and the Growth of Intra-Day Payments

Innovation in the payment system derives from a single but highly consequential phenomenon: the shift from mail and telex to modern information and telecommunications technology. This shift has made it possible to effect payments in real time at a negligible transaction cost, irrespective of the size of the payment, the unit of account, and the location of the traders involved. This means that the velocity of circulation of money can increase enormously with respect to traditional standards. This far-reaching change in the technology of payments has coincided with an exponential increase in the volume of payments, largely as a consequence of the expansion of financial transactions outlined in the previous section; as a result, the annual flow of payments now amounts to several times GDP in the main industrial countries.

For the traditional structure of the payment system and the central bank’s role in it, the combination of these developments has created new problems, possibilities and risks. Most of the recent growth in payment volumes has been accounted for by net settlement systems, such as CHIPS in the United States, CHAPS in the United Kingdom, BOJNET in Japan, and SIPS in Italy. However, since payments flowing through these systems are provisional, as they are settled in monetary base only at the end of the day, the increase in total gross payments relative to net balances has prompted growing concern about settlement risk. In fact, failure to settle by a major participant can set off a chain reaction through the “unwinding” of the day’s transactions. In other words, the pyramiding of provisional intra-day payments involves an ever-increasing systemic risk. In a well-known paper, Humphrey (1986) estimated that on a typical working day the default of a single major participant in CHIPS (which has since been reorganized to reduce systemic risk) would bring down about 40 percent of the other participants. Analogous exercises conducted elsewhere gave similar results.

An alternative route, taken by the Federal Reserve in the early 1980s, is to offer continuous settlement services in central bank money. This has the immense advantage of dramatically reducing systemic risk as the “finality” of payments is provided throughout the working day for the gross amounts that would otherwise go through the clearing, rather than only for the net balances resulting from the clearing itself. However, it also implies a much greater need for central bank money, since the increase in the velocity of circulation permitted by telematics is not sufficient to accommodate the increase in the transaction demand for base money generated by its use for gross settlement. Hence, for the central bank, the problem is providing its own money in ways and amounts that satisfy both payment system and monetary policy requirements. When the electronic “Fedwire” started in 1982, the Federal Reserve offered an unlimited overdraft facility at no cost. Only later—with the explosive growth of intra-day payments—did the problems and risks implicit in this solution become apparent.

The new challenge that central banks are facing in payment systems is thus to find a satisfactory way to combine efficiency, security, and effective monetary control. Whatever solution is adopted, it will have an impact on the way the central bank stays in the market; the way the market is monitored, the way monetary policy is implemented, and the way liquidity is supplied to individual banks are all bound to be affected. In this respect, dematerialization and the growth of intra-day payments are as radical a change compared with paper technology as the latter was compared with commodity money. As many episodes in recent years have shown, the extent of systemic risk has grown considerably. Bank runs used to be thought of as the main threat to financial stability. Under present circumstances, however, it is debatable whether this is still realistic, considering the extensive arrangements for deposit insurance and banking supervision. The high velocity of money today suggests that wholesale payment procedures may well now be the main channel through which instability can spread.

The Internationalization of Exchange

A key feature of traditional central banking, and economic policy in general, was the virtual coextensiveness of the central bank’s jurisdiction with the “market.” Though international trade and finance have always existed, national markets had long remained sufficiently segmented to make political and economic sovereignty but two sides of the same coin. This correspondence is being rapidly eroded. To use a figure of speech, what used to be the “content,” namely the activity of individual economic agents (including banks) within a national system, is outgrowing the “container,” namely the area of competence and authority of the central bank.

As a result of the high international mobility of capital, the removal of exchange controls, and the increasing freedom to offer cross-border financial services, the economic agents involved in a financial transaction, the currency used, and the financial market or intermediary no longer have to carry the same national label. In a regime of perfect capital mobility and no foreign exchange restrictions, firms and households can place themselves beyond the reach not only of their own central bank but also of any central bank acting in isolation. Financial institutions can, to a large extent, choose the regulatory regime they prefer. The notion and practice of arbitrage has considerably widened its scope.

As a result of technical progress, increasing economic integration and, most of all, the internationalization of monetary and financial activity, the volume of cross-border payments—that is, payments involving residents of different countries and/or two different currencies—has increased enormously. Although data on cross-border payment flows are not readily available, the indirect evidence is sufficiently clear. For instance, the volume of payments handled by SWIFT, the main carrier of cross-border payments, has increased by 60 percent over the last five years. The growth of cross-border transactions in securities, as captured by balance of payments statistics, has also been extremely rapid. To cite but one example, in the United States such transactions rose from about 9 percent of GDP in 1980 to more than 100 percent in 1992.

In these circumstances, the traditional functions of central banking are increasingly difficult to perform. For example, it is not clear how much control national monetary authorities retain over their “national” aggregate demand or their “national” price level. One may object that capital mobility is still not perfect and that individual countries may regain some degree of monetary policy independence, even with considerable capital mobility, provided they are prepared to forgo control of the exchange rate. I believe that both these objections are questionable. The 1992 currency crisis showed how quickly capital flows can respond to economic incentives and to shifts in expectations. And events since then have confirmed yet again that the monetary policy independence granted by floating exchange rates may be largely illusory. These are, of course, complex theoretical and empirical questions, which cannot be addressed properly in the context of this paper. For my purposes, what matters is that today national monetary authorities must take into account the “mood” of international investors and the stance of monetary policy in other countries to an extent unparalleled in the past.

Similarly, the rapid growth of cross-border payments is having an enormous impact. Since the pyramidal structure typical of national payment systems does not encompass cross-border payments, the growth of the latter creates a demand for central banking services that is increasingly being met by private institutions. Individual banks and groups of banks provide forms of international service and organization similar to those supplied by central banks within their national systems. However, the reasons that led, historically, to the development of central banking functions at the national level are even more compelling at the international level: systemic risk is likely to be far greater in the international payment system; competitive pressures are stronger; the disparities between banks are more pronounced; the information needed to assess theft standing is harder to obtain; supervisory rules differ appreciably; and there is no well-identified monetary authority. In short, arrangements for cross-border payments may affect the structure of banking and financial markets and have far-reaching implications for supervisory, monetary, and lender-of-last-resort policies.

Adapting Central Banking

The various challenges just described did not all emerge at the same time, nor were they equally strong in all countries. As far as price stability and monetary policy are concerned, the critical years were the 1970s, when the definitive abandonment of convertibility and the strength of wage pressures made it most difficult to provide a stable anchor for the currency. The challenges of technological change and internationalization became stronger in the 1980s.

Responding to the challenges is no rapid, straightforward process. Institutions adapt through trial and error. The problems they are intended to tackle are typically so complex that abstract solutions, thought out in the quiet of the theoretician’s study, seldom work when put into practice. Moreover, in the history of central banking the successful reforms have been those that, while advancing the state of the art to face a new challenge, have at the same time preserved its sound practices and traditions wherever possible. These may be the reasons why pragmatism and conservatism are so often associated with central bankers.

Bearing these factors in mind, I shall now try to show the ways in which central banking has adapted in the last decade or two and continues to adapt to the challenges mentioned.

Institutionalizing the Commitment to Price Stability

If one looks at the statutes of individual central banks—most of which were drafted at a time when gold convertibility was still the basis of the monetary system—it is not easy to get a precise idea of the objective to be pursued by monetary policy. The objective is not specified, or it is described with vague expressions, such as “providing an elastic currency” (the Federal Reserve), “regulating the currency” (the Bank of Japan), “safeguarding the currency” (the Bundesbank), “watching over the currency and credit” (the Banque de France, according to the 1973 Statutes, recently amended), or “regulating the value of the currency in such a way as will be most conducive to the nation’s prosperity and welfare” (the Nederlandsche Bank). The experience of the 1970s and early 1980s convinced many that in a world in which monetary policy had lost any natural anchor, the monetary authorities’ commitment to price stability had to be made unmistakable and credible in the eyes of the public. A strong and explicit commitment to price stability is deemed to be necessary to sustain the public’s confidence in the medium of exchange and, in the end, to preserve a stable currency.

Thus, the response to the challenge of nonconvertibility has been to anchor the central bank to a stated objective and to grant it the requisite independence to resist pressure for easy money. There is little doubt that this evolution has been significantly influenced by the statutes and methods of the central bank of the country with the best postwar record in terms of price stability—namely, Germany.

This evolution has meant two things: making explicit certain rules of conduct that many central banks already follow in practice; and strengthening the autonomy of the central bank in handling monetary policy once the objective has been clearly set. Both theory and practice seem to point in this direction. In this regard, much has been done in a number of countries, most notably New Zealand, Italy, and France. In Europe, the objective of price stability has been given quasi-constitutional status by its explicit inclusion among the provisions of the Treaty of Maastricht.

Having made this premise, two qualifications are necessary. First, even in the new context, the task entrusted to central banks remains inherently complex; the commitment to price stability cannot be pursued to the point of eliminating all flexibility or discretion in the handling of monetary policy. As Paul Volcker put it recently, there is no “quick fix” to establish the credibility of the anti-inflation commitment (Volcker, 1993). In the real world, which, unlike the smoothly functioning economies of abstract models, is marred by imperfections and therefore needs to be governed, excessive rigidity in the interpretation of price stability or reliance on mechanical rules for monetary management could prove to be a cure worse than the disease.

Second, while an autonomous central bank is certainly an important component of a credible commitment to price stability, we must not forget that no degree of autonomy can guarantee this objective if other areas of economic policy are set on divergent paths. Indeed, central bank autonomy offers no escape from the consequences of mismanagement of fiscal and incomes policies.

Redesigning Regulation

In most countries financial regulation has traditionally been based on the division of the financial system into the securities, banking, and insurance sectors, with central banks generally, though not always, responsible for banking supervision. In terms of contracts and institutions, each sector was assumed to be sufficiently separate from the others to require specific regulation with little need for coordination between the various regulatory authorities. As was noted above, financial innovation has weakened the rationale for this approach and appears to have seriously undermined its effectiveness.

In recent years, the adaptation of the regulatory system to the blurring of financial boundaries has followed several routes.

A first route was the gradual abandonment—which is still under way and by no means completed—of the traditional sharp distinction between commercial and investment banking inherited from the 1930s. Even in the United States and Japan, the regulatory barriers between the two segments of finance have been significantly reduced.

A second route has been to move from a purely legal to an economic notion of the supervised institution. The most significant step so far, taken in the mid-1980s and now accepted in many countries, has been the introduction of consolidated supervision of banks. Organizing supervision of financial conglomerates is the much more complex task that supervisors are now trying to accomplish.

A third route has been the drive to establish uniform regulation for each type of business. This can be called a “functional” approach, one in which each function, or type of financial service or product, is subject to a particular supervisor, regardless of the type of financial institution that offers it. This appeared to be an advance with respect to the more traditional “institutional” approach, whereby each category of institution is subject to the control of its own supervisory authority for the entire range of its operations, because it ensures, as the jargon goes, a level playing field.

The functional approach, however, has serious limits of its own, arising from the danger of concentrating on individual parts of a complex reality without understanding the whole: in the end, it is institutions that fail, not the single functions they perform.

In my own country, Italy, we have tried to overcome these problems by organizing the instruments and responsibilities of supervision according to objective rather than by function or institution. The approach is based on the recognition that regulation serves two aims. The first is to foster the efficiency of the market in managing information and pricing financial assets. This is pursued through all forms of intervention regarding the market as such, the rules of conduct for its participants, disclosure requirements, and the setting of contractual standards. The second objective is to preserve stability. Even a well-designed system of standards for disclosure and rules of conduct cannot eliminate all the sources of instability, because some trading takes place outside the market, some agents produce incorrect information, some relevant information is simply not available or retrievable, and so on. Hence, rumors and misleading or false information may distort the allocative process and even lead to financial crises. Financial innovation could actually aggravate this problem, since both the opening of new markets and the process of risk unbundling may shorten agents’ time-horizon, making it harder to evaluate credit risk correctly and accelerating the transmission of local shocks to other parts of the financial system. Stability regulation is therefore needed both to reduce the risk of collapses of the public’s confidence in financial institutions and to limit the consequences of crises that follow such breakdowns.

On the basis of this distinction, Italian regulation has been reshaped in recent years to ensure that each of the two aims falls within the sphere of competence of a single authority: stability regulation has been placed within the realm of the central bank not only for banks but also for securities houses, mutual funds and other financial institutions, while market-making regulation has been assigned to the Consob, the Italian equivalent of the Securities Exchange Commission (SEC).

Reducing Risks in Payment Systems

Reliance on net settlement systems has made it possible to avoid an increase in the holding of non-interest-bearing central bank money parallel to the enormous increase in financial transactions in recent years. The multilateral nature of netting schemes, however, has entailed a concomitant increase in systemic risk. Of course, the central bank always has the capacity to provide enough liquidity to the system to prevent a crisis from erupting. But if the stability and safety of systems were based on the assumption that the central bank will always do so, risks would in the end grow rather than diminish and monetary control would be lost. Outside money would in fact cease to be exogenously determined and the very essence of central banking would be undermined.

Moral hazard and the externalities associated with netting systems therefore require central banks to reassess and enhance their role within the payment system. The response of central banks to these challenges is in fact developing along two lines. At one level, central banks are promoting measures to enhance security in domestic netting systems, using their authority as overseers of the payment system. At another level, they are expanding their operational role, offering more services and developing real-time gross settlement systems (RTGS).

The measures that central banks have developed and promoted to enhance security in domestic netting systems cover a wide range of aspects: restrictions on direct access of supervised credit institutions and public entities to interbank netting systems; the adoption on a voluntary basis of bilateral limits on credit exposures; the capping of multilateral exposures by central banks; the collateralization of debt exposures, and, in the United States and Japan, the stipulation of liquidity-sharing and risk-sharing agreements among participants. In some countries relevant changes in the legal framework have been introduced to ensure the legal validity of bilateral and multilateral netting. In other countries, the pricing of central bank services has been used to smooth the functioning of the clearing and settlement procedure.

The second response, a shift towards continuous settlement in central bank money in the course of the business day, has become a central element in the evolution of modem interbank payment systems. In the United States, Japan, Switzerland, Denmark, and other countries, such systems are already in place. In most other advanced countries they are being developed. For commercial banks, switching to RTGS entails the cost of increasing their reserve holding, which explains why they are somewhat reluctant. In fact, although technological progress tends to increase the velocity of central bank money dramatically, the shift from net to gross settlement is likely to mean an increase in the level of both transactional and precautionary reserves. A factor that has favored acceptance of RTGS is the rising costs that participants in netting schemes have to bear as a result of risk reduction measures.

It is difficult to foresee the precise impact that the shift to gross settlement will have on the demand function for central bank money. Other things being equal, the outcome depends on two factors. The first is whether gross systems will supplement netting systems or replace them. As replacement may be brought about by a decision of the central bank, this is a hotly-debated issue in several countries in Europe and elsewhere. I personally believe that the coexistence of the two systems is preferable. In fact, if the risk reduction measures of netting schemes were such that the two systems entailed equal risks, then payment operations would tend to be distributed between them in such a way as to minimize their cost. And it is reasonable to assume that, for certain classes of payments, netting systems will continue to be the efficient mode.

Second, the final outcome will also depend on the ways and conditions in which central bank money is supplied. All central banks face today what can be called the problem of liquidity provision to gross systems. Queuing, compulsory reserves, overdrafts, and collateral are the constituent elements of a complex architecture, still to be designed in detail. The interplay of factors makes it difficult in the extreme to design the perfect architecture of a gross or, more generally, a large-value payment system. Once more, a degree of pragmatism is necessary. In particular, central banks should be prepared to bear part of the higher costs associated with gross settlement.

Whatever choice a central bank makes on these matters, changes occurring within the payment system require adaptation of the procedures by which the central bank supplies money to the banking system and have major implications for the operating procedures of monetary policy.

Enhancing Central Bank Cooperation

As noted above, the increasing integration of world capital markets undermines the ability of the single country to pursue purely domestic goals through monetary policy. That there is an internal inconsistency in the coexistence of free circulation of goods and services, free capital mobility, fixed exchange rates, and independent monetary policy is by now a commonplace idea, although it took time to arrive at it. The Bretton Woods system collapsed, and more recently the European Monetary System was seriously shaken, because individual countries failed to acknowledge their inconsistency and accept the requisite degree of policy coordination.

Cooperation in the area of monetary policy is clearly important but, one might argue, not indispensable, if countries are willing to accept one of the two following solutions to the coordination problem: surrendering monetary sovereignty altogether, by “mimicking” the monetary policy of a larger partner; and letting the exchange rate fluctuate freely. Both these routes, however, pose problems. Pegging to a larger currency implies acceptance of a policy stance designed for the needs of another economy and renouncing all say in the decision-making. Floating means leaving the decision to markets, which have repeatedly shown their inability to avoid bubbles, overshooting, and volatility that are detrimental to economic activity. Over this century international monetary policy moved continuously between the three poles of coordination, pegging, and floating. Recently, the European Union has adopted the radical solution of a single currency and a single central bank and enshrined it in a Treaty. The test of implementation is for the years to come.

Monetary policy was long the only central banking function for which the game of international cooperation was played, albeit often in words rather than substance. Cooperation is much more recent with respect to the other two functions of the triad—about twenty years for banking supervision and ten for payments systems—but it has made more progress than in the monetary policy area. As regards supervision, the internationalization of finance compels national authorities to seek greater harmonization and transparency of prudential standards both to safeguard the stability of banks and to ensure fair competition. The Basle Committee on Banking Supervision (once known as the Cooke Committee) has risen, over the years, to the role of key rule-making body. Its pronouncements have covered such areas as capital ratios, risk concentration, standards for banking supervision, and consolidated supervision. It is currently working on market risks, financial conglomerates, and derivatives.

No less serious a threat to financial stability comes from the development of international payment services. Here, very little can be accomplished without a cooperative attitude, as no central bank acting on its own can offer viable solutions to the problems examined in this paper. In 1988 I argued that “the choice that is confronting central banks is whether and how much to get involved in the developments that will lead to satisfying this need. The options range from directly and jointly operating a clearinghouse, possibly through an institution jointly owned by central banks, to fully delegating these functions to the private sector” (Padoa-Schioppa, 1988b). Subsequent events have only strengthened my conviction. At the same time, central bank cooperation has made significant progress and produced concrete results. The Group of Experts on Payment Systems, which in 1990 drafted the so-called Lamfalussy Report, and the Group of Ten Committee on Payment and Settlement Systems that succeeded it, have made considerable headway in this complex but crucial matter. The latter’s recent so-called Noel Report on central bank payment and settlement services with respect to cross-border and multi-currency transactions, in particular, provides a detailed account of available options, including, as was anticipated in the passage quoted above, the possibility of setting up a Common Agent of Group of Ten central banks entrusted with settling cross-border, multi-currency payments on its own accounts.


The history of central banks is one of ongoing adaptation of a basic mission (governing money) to new challenges posed by changes in the technological, economic, and institutional environment. When central banking began about three centuries ago, the birth of the modern state posed the problem of ensuring the government rapid and effective access to external finance in order to face emergencies such as war and famine. At the beginning of the nineteenth century, the challenge raised by the spread of bank notes was addressed by imposing statutory rules on the issuing banks and, in many countries, by centralizing the note issue. Later in the century, the unprecedented spread of deposit banking required a new type of adaptation, which took the form of lending of last resort and banking supervision. More recently, in our own century, the need to manage money for purposes of macroeconomic stabilization led the state to get a firmer hold on central banks, in many cases through nationalization.

The challenges confronting us today can be grouped under four headings: (1) preventing the pressures of labor and governments for money creation from jeopardizing price stability; (2) revising the regulatory structure, to ensure financial stability in the face of increasingly complex contracts and markets; (3) reshaping the payment system with the shift from paper to computerized information technology; and (4) extending the three central banking functions to the international economy to keep pace with global and integrated markets. The responses to these challenges, to judge from current developments, can be summarized as follows.

Progress has been made with respect to the first challenge, by emphasizing the objectives to be pursued by the central bank and strengthening central banks’ autonomy. Very broad central bank autonomy, with due emphasis on price stability as its primary objective, is by now an accomplished fact in many industrial countries.

Signs of response to the second challenge are also visible. Indeed, central banks and policymakers around the world increasingly recognize that they can no longer restrict their attention exclusively to the banking system, if overall financial stability is to be preserved.

In the field of payment systems, an area that after many years of relative neglect now ranks high again in the priorities of central banks, measures to reduce risks in netting systems and the development of gross settlement are the two main lines of response.

Globalization poses a challenge of a different nature. Individual central banks owe their legitimacy, to use Stiglitz’s (1989) technology, to their “universality” and “coercive powers”—properties they derive from being part of the machinery of the state. This is why monetary sovereignty has always evolved in close association with political sovereignty. In a world still composed of separate political entities, but increasingly unified in the economic and financial fields, how can an effective structure of monetary governance be ensured? While Europe has taken the bold step of planning the creation of a central bank ahead of full-fledged political union, on a global scale the only conceivably feasible route is voluntary cooperation among sovereign national institutions: a second-best route on which considerable progress has been made.


    AngeliniPaolo and FrancoPassacantando“Central Banks’ Role in the Payment System and its Relationship with Banking Supervision,”Giornale degli Economisti e Annali di EconomiaVol. 51 (September–December1992) pp. 45391.

    • Search Google Scholar
    • Export Citation

    CameronRondoA Concise Economic History of the World: From Paleolithic Times to the Present (New York: Oxford University Press2nded.1993).

    • Search Google Scholar
    • Export Citation

    de CeccoM.“New Forms of Financial Regulation and the Evolution of Financial Firms,”The International SpectatorNo. 2 (April-June1993) pp. 519.

    • Search Google Scholar
    • Export Citation

    FisherIrvingThe Purchasing Power of Money: Its Determination and Relation to Credit Interest and Crises 2nd ed. (New York: Macmillan, 1920). Originally published in 1911.

    • Search Google Scholar
    • Export Citation

    Folkerts-LandauDavid and PeterGarber“The ECB: A Bank or a Monetary Policy Rule?,”in Establishing a Central Bank: Issues in Europe and Lessons from the USed. by MatthewB. CanzonieriVittorioGrilli and PaulR. Masson (Cambridge, England, and New York: Cambridge University Press1992).

    • Search Google Scholar
    • Export Citation

    GianniniCurzio“Confidence Costs and the Institutional Genesis of Central Banks,”in Temi di Discussione del Servizio Studi della Banca d’ItaliaNo. 226 (May1994).

    • Search Google Scholar
    • Export Citation

    HayekFriedrichA. vonMonetary Nationalism and International Stability (New York: A.M. Kelley1964reprint of 1939 edition).

    HicksJohnR.“A Suggestion for Simplifying the Theory of Money,”EconomicaVol. 2 (February1935) pp. 119.

    HicksJohnR.(1967a) “The Two Triads,”in Critical Essays in Monetary Theory (Oxford: Clarendon Press, 1967).

    HicksJohnR.(1967b) “Thornton’s Paper Credit (1802),”in Critical Essays in Monetary Theory (Oxford: Clarendon Press1967).

    HirschFred“The Bagehot Problem,”The Manchester School of Economic and Social StudiesVol. 45No. 3 (1977).

    HumphreyD.B.“Payments Finality and the Risk of Settlement Failure,”in Technology and the Regulation of Financial Markets: Securities Futures and Bankinged. by AnthonySaunders and Lawrence J.White (Lexington, Massachusetts: Lexington Books1986).

    • Search Google Scholar
    • Export Citation

    Padoa-SchioppaTommaso (1988a) “The Blurring of Financial Frontiers: In Search of an Order” (paper presented at the Conference on Financial Conglomerates, Brussels1988) pp. 1415.

    • Search Google Scholar
    • Export Citation

    Padoa-SchioppaTommaso (1988b) “International Payment Systems: the Function Begets the Organ” (paper presented at the International Symposium on Banking and Payment Services Board of Governors of the Federal Reserve SystemWashington, D.C.June 91988).

    • Search Google Scholar
    • Export Citation

    Padoa-SchioppaTommaso“Verso quale banca centrale?,”in La moneta e il Sistema dei Pagamenti (Bologna: II Mulino, 1992).

    StiglitzJosephE.“On the Economic Role of the State,”in The Economic Role of the Stateed. by ArnoldHeertje (Oxford and New York: Basil Blackwell, 1989).

    • Search Google Scholar
    • Export Citation

    VolckerPaulA.“The Independence of Central Banks: Its Value and its Limits” (address for the Centenary of the Bank of Italy, Rome, December 11, 1993).

    • Search Google Scholar
    • Export Citation

The author is grateful to Curzio Giannini for his kind assistance and helpful comments in the preparation of this paper. The opinions it offers are those of the author, as is the responsibility for any errors.

    Other Resources Citing This Publication