Tijdschrift voor
ISSN 0772-7674
Driemaandelijks publikatie van de Faculteit der Economische en Toegepaste Economische Wetenschappen van de Katholieke Universiteit Leuven, uitgegeven met medewerking van Ekonomika, vereniging van afgestudeerden van de faculteit, uitgegeven met de steun van het Ministerie van de Vlaamse Gemeenschap Departement Onderwijs. REDACTIE: Hoofdredacteur: Prof. Dr. P Sercu Kernredactie: de Professoren F. Abraham, G. De Bruyne, Z. Degraeve, P De Grauwe, M. Dekimpe, D. Heremans, C. Lefebvre, F. Spinnewyn, C. Van Hulle, R. Veugelers en Mevr. A. Gaeremynck en de Heer H. Dewachter. Redactieraad: Deze raad omvat naast de hoofdredacteur en de leden van de kernredactie eveneens: professoren P Beghin (R.U.Gent), H. Daems (K.U. Leuven), R. De Bondt (K.U.Leuven), M. Dombrecht, (Nat. Bank, Brussel), S. Proost (K.U.Leuven), E. Schokkaert (K.U.Leuven), P Van Cayseele (K.U.Leuven) W. Vanthielen (L.U.C. Limburg), J. Vuchelen (V.U. Brussel). Redactiesecretariaat: A. Ronsmans Tijdschrift voor Economie en Management Naamsectraat 69, 3000 LEUVEN Tel. 016J32.66.88 - Fax. 016l32.66.10 - 32.66.98 E-mail:
[email protected] Verantwoordelijke uitgever: P. Sercu, Stosbergstraat 3 , 3052 Oud-Heverlee (Blanden) De Redactie beoordeelt de kwaliteit van de bijdragen maar kan niet verantwoordelijk gesteld worden voor de inhoud. The Editorial Board judges the quality of the contributions published but can not be held responsible for their content. ABONNEMENTSVOORWAARDEN: Volledige jaargang: 4 afleveringen (ca. 500 blz.) 2.000 BF als steunabonnement* 1.200 BF als institutioneel abonnement (bibliotheken, instellingen, bedrijven) 800 BF als persoonlijk abonnement (vooraf te betalen via persoonlijke rekening) 500 BF als student 1.500 BF of $40 als buitenlands abonnement 350 BF als los nummer Abonnementen en bestellingen worden uitsluitend vereffend op PR. 000-01 12553-33 Tijdschrift voor Economie en Management, Leuven.
KATHOLIEKE UNIVERSITEIT TE LEUVEN FACULTEIT DER ECONOMISCHE EN TOEGEPASTE ECONOMISCHE WETENSCHAPPEN
economie en The Financial Sector in Europe: Changes and Adjustments
he ma nummer INHOUD Advertentie Voorwoord F. ABRAHAM The Consequences of European Financial integration for the Regulatory Authorities M. ONADO
231 233
Competition and Cooperation among European Exchanges: the Impact of Financial Globalisation with Special Referentes to Smaller-Sized Exchanges J.-!? ABRAHAM EN A.-F. PIRARD
261
Principles of Corporate Governance with an Application to the Financial Sector C. VAN HULLE
299
Traditionele versus optie-waarderingsmethodes voor groeibedrijven: er zijn geen mirakeloplossingen F! SERCUEN C. VAN HULLE
323
Advertenties The Banking on the Eve of the 21st Century H. VERWILST Boekbesprekingen Eindverhandelingen
SEPTEMBER 1999 DRIEMAANDELIJKS
341 355 359
JAARGANG XLIV
bestendig.3
Generale Bank. Mijn Kracht.
Tijdschrift voor Economie en Management Vol. XLIV, 3,1999
Als hoofdredacteur ben ik bijzonder verheugd onze lezers dit themanummer over The Financial Sector in Europe: Changes and Adjustments aan te bieden. Het onderwerp is ongetwijfeld relevant. Niemand kan blind blijven voor de snelle ontwikkelingen van de financiële sector. Het Europese financiële landschap is in korte tijd grondig hertekend. Het forumartikel van Herman Venvilst evalueert de banksector op de drempel van het nieuwe millenium. Op systematische wijze overloopt de auteur de drijfveren die de banken in België en Europa tot verandering dwingen. Hij hecht hierbij veel belang aan de creatie binnen Europa van een ééngemaakte financiële markt met eenheidsmunt alsook aan de doorslaggevende rol van de technologie in het bankgebeuren. Deze inzichten laten hem toe toekomstige kansen en uitdagingen voor de banken te identificeren. In zijn bijdrage tot dit nummer benadert Marco Onado de veranderingen in de financiële sector vanuit het standpunt van de regulerende autoriteiten. Hij schetst de evolutie van de financiële regelgeving in Europa en legt hierbij de nadruk op het proces van financiële integratie in de Europese Unie. Vervolgens gaat hij na hoe het huidige regelgevend kader kan worden aangepast aan de Europese realiteit van dit ogenblik. Jean-Paul Abraham en Anne-Française Pirard richten de schijnwerpers op de Europese aandelenbeurzen. Onder druk van dezelfde oorzaken die reeds in de vorige bijdragen aan bod kwamen, concurreren de Europese beurzen onderling steeds meer en gaan zij tegelijkertijd op zoek naar mogelijke samenwerkingsverbanden. Volgens de auteurs is samenwerking de enige mogelijke overlevingsstrategie voor kleinere aandelenbeurzen op langere termijn. Cynthia Van Hulle verlegt het debat naar de ondernemingsdimensie en meer bepaald naar de beslissingsstructuur binnen de onderneming. Zij plaatst de verschillende modellen van corporate governance in perspectief en situeert de Belgische positie in dit verband. Vervolgens bespreekt de auteur de problematiek van corporate governance vanuit de recente ontwikkelingen in de Belgische en Europese financiële sector.
De laatste bijdrage is van eerder financieel-technische aard maar is zeer actueel in het licht van de sterke hausse in beurkoersen en de vele introducties van nieuwe aandelen. Piet Sercu en Cynthia Van Hulle vergelijken traditionele en optie-waarderingsmethodes voor groeibedrijven. Zij tonen aan dat er geen mirakeloplossingen zijn voor de correcte waardering van ondernemingen. Meerdere bijdragen in dit themanummer komen voort uit de Fortis Bank Leerstoel (voorheen Generale Bank Leerstoel) die elk jaar gezamenlijk door de Faculteit Economische en Toegepaste Economische Wetenschappen en de Rechtfaculteit van de KU Leuven wordt ingericht. Gedurende het academiejaar 1998-1999 handelde de leerstoel over Financial Services and Financial Markets in Europe: Changes and Adjustrnents. Vanwege het Tijdschrift en Management wens ik Fortis Bank hartelijk te danken voor de financiële steun bij de uitgave van dit themanummer. Aan alle de lezers van het Tijdschrift wens ik alvast veel boeiend leesgenot toe. Prof. Filip Abraham Hoofdredacteur
Tijdschrift voor Eco~iomieen Management Vol. XLIV, 3,1999
The Consequences of European Financial Integration for the Regulatory Authorities by M. ONADO*
I. INTRODUCTION The objective of this paper is twofold. First, to highlight the main characteristics of a 20-year process of European integration hinged on regulatory harmonisation through Directives and monetary union. Second, to examine whether the present regulatory structure, both at the European and national level, is adequate to achieve the four final objectives of regulation: price stability; solvency of financial institutions; disclosure and protection of investors; competition. The paper is organised as follows. Section I1 briefly reviews the rationale and final objectives of financial regulation. Section I11 reviews the evolution of financial regulation in the industrialised countries and the effort of European harmonisation. Section IV analyses the European strategy of achieving a single market in financial services and tries to measure the results achieved so far. Section V points out the steps that should be made, mainly at the European level, to reconcile financial integration with investor protection within an European regulatory framework. 11. THE RATIONALE FOR FINANCIAL REGULATION The raison d'êtue of financial regulation, as in any form of economic activity, is to correct market failures. Both economic theory and his* Uiliversity of Bologna, ILaly
torical experience show that financial intermediation is particularly prone to market failures, due to information asymmetries that dominate the relationship between borrowers and lenders. Tlie asymmetries can destroy the fiduciary relationship that lies at the heart of both using money and lending. This is why the question of whether financial intermediation needs to be regulated is closely linked to the question of why financial intermediation exists (Onado (1992)). Both from a practica1 and a theoretica1 point of view, the debate on financial regulation started from banks. Bank liabilities (sight deposits) are generally accepted as a means of payment, while bank assets are represented by risky loans whose maturity is longer than liabilities. This leads to a liquidation value of bank assets lower than liabilities value; hence to a risk of run and contagion. This is sufficient to explain why we need public intervention to: i) control the money supply and via price stability the fiduciary nature of bank-created money (hence monetary policy and the macroeconomic function of centra1 banks); ii) control the liquidity and solvency of bank to protect depositors (hence, supervisory powers). Credit of last resort granted by centra1 banks is the link between the two functions. As to solvency regulation, some scholars focus on the problem of protection, particularly of smal1 depositors who do not have either the incentive nor the competence to collect information to intervene into bank management (Dawatripont-Tirole (1996)). Another school of thought points to the strict link between the monetary policy function and the supervisory function and stresses that historically the latter led to the former and not viceversa (Goodhart (1985)). The first approach completely separates the problem of bank regulation from centra1 banks and monetary policy and assigns to bank regulation the objective of bank stability (or bank solvency). The latter identifies two objectives: monetary stability and bank solvency which are complementary and strongly linked together. Although bank stability has been historically the first economic need that regulation was meant to satisfy, there are other motives for regulating financial markets. Al1 financial contracts are dominated by information asymmetries and require to protect investors against insufficient information, fraud,
negligence, conflicts of interest, mismanagement etc. These asymmetries can undermine the operations of the markets, because they make it impossible to investors to assess the 'real' quality of firms. The objectives of regulation are therefore transparency of information and correctness of behaviour. A specific side of this problem bears on the control function (i.e. the function of distributing residual claims on the firm's assets) and the market for corporate control. Regulation must draw the general framework of duties and rights of the shareholders involved, managers, and lenders. The rules generally known as corporate governance are therefore an important component of this branch of regulation. Finally, there is another motive why economic activity, and financial marltets in particular, need to be regulated, i.e. to remove obstacles to competition which is essential to productive and allocative efficiency. Financial markets may be unable to reach autonomously competitive conditions due to the specific nature of essential services (in the payment system there are natura1 monopolies such as information networks) or to information asymmetries that make retail banking markets contestable only to a limited extent. We have thus identified four fundamental theoretica1 motives for bank and financial regulation: price stability, bank (and other intermediaries) stability; disclosure (to use a single word); competitionl. The pendulum of actual regulation swings between these four positions, according to the evolution of historica1 conditions and the political motives driving regulatory action. The instruments to achieve these objectives, whose strength and scope vary accordingly, are: 1. traditional instruments of monetary policy (reserve requirements? discount window and open market operations); 2. lending of last resort; 3. insurance funds and compensation schemes; 4. prudential regulation instruments which focus on the solvency and stability of financial institutions and are based on three main lines 4a) structural controls (interest rate controls; entry restrictions; lines of business restrictions); 4b) minimum requirements (capita1requirements for credit risks, market risks etc.; reserve requirements);
5.
6. 7.
8.
4c) evaluation of fit and proper conditions (both at the moment of the authorisation and during the lifetime of the financial firm); conduct of business regulation. These controls focus upon the relationship between intermediaries and their clients2; disclosure of information by issuers and corporate governance rules; interventions to maintain competitive conditions (approval of mergers; ban of restrictive business practices etc.). special procedures in case of insolvency.
The institutions which are committed to the achievement of the final objectives, using these instruments are: 1. centra1 banks; 2. bank regulatory agencies; 3. securities regulatory agencies; 4. insurance regulatory agencies; 5. antitrust authorities.
The balance between these four objectives, eight classes of instruments and five regulators undenvent a dramatic change in the last decades, as a response to different macroeconomic conditions and growing interdependence and integration of national economies. Europe is a significant case in point, owing to the driving force of the efforts to achieve a Single Market. 111. EUROPEAN FINANCIAL REGULATION: THE STYLIZED FACTS A. The situation at the dawn of Europe
When the Treaty of Rome was signed, the financial regulation of the six founding countries involved - as the underlying economic conditions - were very different. However, in al1 countries the regulation prevailing in the mid-50s was the offspring of the nightmare of the Great Depression. The financial crises prompted in al1 countries deep reforms of bank and (at least in the US) financial markets. For the first time in economic history different countries were trying to work out a solution to the biggest "market failure" that ever occurred. Although stemming from the
same problem, financial reforms were still national in scope, mirroring the deep differences in the underlying economic and institutional conditions and the low degree of interdependence. The only common characteristic was the heavy (and to some extent new) weight given to the objective of bank stability and to the function of monetary policy as a macro-economic tool. This meant that lesser emphasis was given to the other two objectives of financial regulation: disclosure and competition. The hypotheses underlying the emphasis given to bank stability can be thus summarised: i) ii) iii) iv)
v)
banks are more prone to liquidity and solvency crises; a bank crisis is likely to assume a systemic dimension; bank depositors are particularly worth of protection; savings non invested in bank deposits is less worthy of protection. The caveat emptor principle should apply. The protection stems therefore from ordinary civil law, property rights rather than from special legislation; in the banking markets competition can increase the propensity to risk and hence financial fragility.
While the first two hypotheses are still part of the received wisdom in banking theory, the remaining three are fundamentally a mirror of the specific conditions of those times, namely the fear of (and lack of specific instruments against) bank instability on the one hand and the overwhelming importance of banks in the financial systern on the other. As a matter of fact, the Great Depression was the last of a long chain of bank crises, dating back to the creation of commercial banks in the present form. Both practitioners and economists soon discovered the high probability of bank crises and found the underlying causes in three characteristics of the new-borne institutions. The liabilities were redeemable at sight and circulated as a means of payment; the cash reserves were a fraction of the deposits; the interest-bearing assets had a longer duration than deposits and were to a great extent represented by risky loans, subject to those which wil1 be later called information asymmetries. Although many authors (Thornton, Bagehot) gave a thorough theoretical analysis of the problems involved, the instruments used were mainly confined to lending of last resort and the "art" of centra1banking.
At the eve of the bank crises of the '30s' the situation had hardly changed. After the terrible shock of those years, regulators were particularly concerned of stability but they also knew that the instruments they should rely upon had hardly changed since Bagehot's times. This is the main reason why they decided to sacrifice on the altar of stability both disclosure and competition. The concept of the 'unconscious saver' was the key for putting the problem of disclosure (and protection of investors) wel1 behind stability. The typical buyer of a security was considered to be a "conscious" investor fully able to assess the risks involved and therefore entitled to receive available information. So deep was this belief that most countries did not find it necessary to issue special legislation on this point. Only in the United States the Banking and Securities Act were issued simultaneously. Moreover, in the climate of opinion prevailing in the '30s7particularly in European countries, regulators looked with a suspicious eye to competition. Hence, the fifth hypothesis. Although it is true that many banking crises were characterised by overbanking and cut-throat competition, it is also true that in those times there was no specific instrument to monitor bank risks. In many respects, the regulators of the '30s were inclined and to some extent forced to think that in the future restraining competition would help to keep bank rates to levels favourable to bank profitability. If oligopolistic conditions followed, it seemed a fair price to pay for soundness and stability. This is why bank regulation was characterised by heavy structural controls, barriers to entry, restraints to assets and liabilities. The whole fabric of financial regulation was designed to ensure that institutions remained profitable most of the time. The "uniqueness" of banks from a theoretica1 point of view was the justification for the special regulation and the extensiveness of structural controls. In most European countries, there was a further motive: the myths of economic plan and public ownership of banks went hand in hand in those times, giving strength to the idea that banking was avery special sector, where the principles of market freedom had somehow to bow to "superior" interests of controlling the economy and the private sector. The net result was a drastic dwarfing of competition both in the market for financial services and in the market for corporate control. Both the suspicion against competition and thepenchant for the con-
trol of credit by the State, explain why in most European countries, bank legislation empowered authorities (centra1 banks arzd the Ministry of Finance) mainly with structural controls. The growing role of monetary policy as an anti-inflation and countercyclical weapon, gave new life to centra1 banks, giving them a key role in the making of the economic policy as a whole. Although always important, central banks became policy makers only after the reforms following the Great Depression. This is one of the reasons why in most European countries the authority in charge of bank stability was an independent commission separated from the central bank. The six founding fathers of the Treaty of Rome adopted the separation solution; Italy was the exception (Padoa-Schioppa (1999)). To sum up, the regulatory framework after the Second World War and at the very moment of the signing of the Treaty of Rome was:
0
discovering monetary policy as an anti-inflation weapon and centra1 banks as policy makers; primarily oriented to the objective of bank stability. Lacking specific instruments, this objective was pursued mainly with discretionary, structural controls. focused on banks; lacking a specific regulation to protect investors in securities (the United States being the only country to innovate in this field); leaning on stability at the expense of competition; exercised by a bank authority, normally separated from the central bank responsible for monetary policy.
B. The evolution of financial regulationfiom the Treaty of Rome to the EMU Four driving forces common to al1 industrialised countries can be found behind the evolution of financial regulation in al1 countries and specifically in Europe in the last decades:
e
the gradual shift of households' portfolios towards securities (particularly, risky securities) issued by the private sector; internationalisation (European integration being a component of a general trend);
mounting politica1 pressure to increase the efficiency of the banking sector; dismal results of the public ownership of banks and administrative controls on credit. Gradually, the necessity of achieving a balanced result among the four objectives has been affirmed in clearer and clearer terms. The growing importance of securities markets coupled with some significant cases of heavy damages to investors brought gradually to the creation of specific regulation to protect investors. France (1967) and Italy (1974) gave the lead; other countries followed suit. Disclosure and rules of conduct were given an outstanding position as regulatory objectives. Ad hoc regulator, more or less modelled on the American Sec were established. The objective of bank stability was more and more pursued through specific instruments instead of through restricting competition and raising ceterisparibus bank profits. Structural controls were gradually phased out in favour of prudential, non-discretionary measures. Gradually since the late '60s, competition in banlcing was more and more seen as an independent objective, as it became clear that the excess weight given to stability came at the expense of the efficiency and benefits to consumers which are the typical outcomes of competitive conditions. Although gradually (and in some countries very slowly) the public policy objectives were achieved by relying upon competition and market forces rather than upon regulation (OECD (1998)). This evolution meant putting aside structural controls in favour of prudential controls. The real watershed has been the international agreement on minimum capita1 requirements (Basle Agreement). Since then, the control of bank stability relies on instruments aiming at assuring a cushion of bank capita1 related to the risks taken by each institution. For the first time bank regulators were in the position of using specific instruments of prudential, ex ante supervision to control the riskiness of bank portfolios. From this point of view, this achievement marks the end of the Bagehot's era. At the same time (but als0 as a consequence of the adoption of capital adequacy measures) this change enhanced the role of market forces in achieving bank efficiency. The adoption of capita1 ratios strengthened the relationship between profitability, growth and equity. Banks had to adopt the same behaviour of other firms and had to compete in the market for products and financial resources. Competition and
stability were no more seen (at least officially) as mutually exclusive (Ciocca (1998)). In European countries the process of regulatory change was also driven by European Directives. It is important to remember that the Directives find their basis in the Treaty of Rome (art. 85 and 86) although only in 1981, with the Court of Justice decision on the Zuchner case, it was explicitly stated that those articles could be applied also to banks (Ghezzi-Magnani (1998)). This means that the idea of the "uniqueness" of banks was wel1 rooted in the European regulators' mind. Both in the Commission and in most countries this was a major cause of tiresome discussion on whether and how design and implement banking directives. It is the main reason why it took twenty years to issue the First Banking Directives and other twelve years to implement it in the country (Italy) which chose the longest possible delay. The E C legislative framework for financial markets is grounded in a concept known as 'competition among rules7which takes the reality of separate and distinct national legal and regulatory systems as given, but states that member State must recognise the validity of each other's laws, regulations, and standards. The mutual recognition was meant to facilitate free trade in goods and services without the need for prior harmonisation. Directly derived from this principle is the Second Banking Directive provision for a single passport, under which credit institutions incorporated in any Member State are permitted to carry out a full range of 'passported services' detailed in the Directive's annex, throughout the Community. Similar guidelines are laid down for the provision of cross-border investment services in the Investment Services Directive. Mutual recognition was built on two pillars: harmonisation of minimum standards of supervision and home country control. The former was meant to provide a common denominator of prudential controls: in other words to restrain the scope for competition among rules above this bottom line. Home country control seemed from the practica1 point of view the only criterion compatible with internationalisation of financial activity and the need for supervision on a consolidated basis. The European convergence made it easier to remove restrictions to financial activities which means adopting the universa1 banking model. A major obstacle to competition within national markets and
among national markets was therefore removed. Those principles were applied to al1 financial activities (although insurance companies succeeded in remaining partially unaffected) so the net result was - at least theoretically - a competition a tout azimut: among intermediaries, among markets, among financial systems. C. Division of responsibilities When the stage becomes larger and the number of players (regulators) keeps growing, the problem of division of responsibilities arises. At the beginning, the division of labour among regulators was organised "vertically" i.e. according to the subject involved: banks, market intermediaries, insurance companies. This solution can reasonably work only when lines of business are rigidly separated by regulation, i.e. by what we have called structural controls, which was the typical situation of the '50s and the '60s. When structural controls were gradually phased out, competition at the border line between banks and nonbank intermediaries, among various categories of banks became one of the main driving force both of market conditions and regulation. The demand for "levelling off the playing field" prompted major changes in regulation, whose main characteristic was the gradual enlargement of activities permitted to individual institutions. But if "vertical" divisions of labour among intermediaries did not work any longer, als0 the division of labour among authorities was more and more questioned. The biunivocal relationship between a given intermediary and its regulator was no longer possible. This is why where major regulatory reforms occurred, regulation criterion, i.e. according was redesigned according to a "li~rizontal'~ to the objective to be pursued. Italy has been one of the first country to implement this principle. In an important piece of legislation passed in 1991, the responsibility for stability (both for banks and investment firms) was assigned to the Bank of Italy, while the responsibility for disclosure and transparency (both for banks and investment firms) was assigned to the Consob. At the Same time, not surprisingly, the separation between monetary policy functions and supervisory functions became a major issue. No country adopting the separation solution reversed this choice, while others began to discuss major changes in bank regulation where independent agencies (deposit insurer etc.) were given new powers.
It is significant that while the institutional independence of centra1 banks as monetary policy makers were reinforced, most Governments widened the space between monetary policy and bank supervision. In other words, they strengthened the separation between macro- and micro-economic functions of centra1 banks. Also in this case, Italy was a major exception, as the union of monetary policy and bank supervision functions was never questioned. The most important change in the European regulatory framework has been a gradual shift of the division of responsibility from subjects (banks VS investment firms) towards a division by objectives (mainly stability versus disclosure!. As the Cob7schairman put it: "The information disclosed to the public and the integrity of markets are of equal importance whosoever the intermediary concerned (bank, investment firm, insurance company) and for any financial product whatsoever (loan, security, insurance ~ o n t r a c t ) " ~ . It is interesting to note that the first document to raise this question was an Australian official report (Wallis Report). This analysis moved from the need to organise supervision by objectives and recommended a redistribution of existing responsibilities between two new authorities, one for bank stability, the other for investor protection. It also argued that the authority responsible for intermediaries' stability must be separated from the centra1 bank. In other words, the latter should be responsible solely for ensuring macroeconomic stability, while microeconomic stability should be pursued by an authority fully responsible of disclosure and rules of conduct. The solution was therefore to divide responsibilities according to the objective to be achieved: stability on the one hand; disclosure on the other. It was therefore called the "Twin Peak solution". Other European countries went beyond this solution and devised a reform based on a single financial services regulator. First the Scandinavian countries, then the UK (Briault (1999)). The British reform announced just after Labour's election victory in 1997, stems from both the dissatisfaction with the existing system and the desire to increase the competitiveness of the London market. The regulatory structure is based on two pillars: a clear definition of tasks and responsibilities of supervision and a unification of overall supervision in a single authority. The reform stresses the need to avoid any duplication of responsibilities to reduce costs of regulation (for a given level of overall efficiency) and enhance the competitiveness of the UK financial community (FSA (1997); HM Treasury (1998)). The
Bank of England will have the responsibility for macroeconomic stability while the new FSA will be responsible for both the objectives of stability and disclosure and in particular for we have called prudential regulation and conduct of business regulation. The FSA will therefore incorporate the responsibilities of the previous supervisory body, the self-regulatory organisations for the various financial industry sectors (9 bodies) and the microstability functions of the Bank of England. Although the reform is still under way, there is a growing consensus in the UK that the single regulator solution is better than a structure based primarily upon the objectives of regulation (Briault (1999)). To sum up, we can say that while the regulatory framework has been greatly changed, particularly in the European countries, the architecture of regulatory authorities - although subject to increasing debate - is only at the first stage of its evolution.
IV THE LONG AND WINDING ROAD T 0 THE SINGLE MARKET The European directives gave a major contribution to the harmonisation of the regulatory framework in al1 member States. Thanks to the common identification of the services admitted to mutual recognition, the genera1 model of the intermediary embodied in each national legislation is now the Same. This model allows each institution to move freely along the whole gamut of banking and financial services. This objective was however an intermediate objective to the fina1 target of realising the single market for financial services. How far are we from this target? If we expect that in a single market prices (interest rates) should be levelled off and quantities ( households' financial assets and firms' liabilities) should reflect only underlying economic fundamentals, it is only too easy to say that the promised land is still in the offing. Three points must be stressed here: i) initia1 conditions of individual member states are very different as they reflect historica1 structural differences of each economy and the regulatory framework as a whole; ii) most legal and fiscal barriers are still in place. Suffice it to mention the difficulties that the present Commission has found in working out a plan for fiscal harmonisation;
iii) financial and particularly banking markets are contestable only to a limited extent. Information asymmetries (mainly in lending) and sunk costs give strength to the position of the incumbent and make it very difficult for the new entrant to build up significant market shares from scratch. The removal of barriers to entry and harmonisation of legislation are therefore a necessary but not sufficient condition for the actual integration of financial markets. The main result of the new European regulatory framework has been an increase in competitive conditions within each national market and among national agents, as a consequence of the deregulation. Not surprisingly the effects have been immediately felt where (i.e. Italy) previous regulation made an extensive use of structural controls such as entry restrictions and line of business restrictions. At the Same time, each financial system has seen a significant increase in the number of mergers which have substantially raised the traditional measures of concentration. This seemingly counterintuitive result has raised more than an eyebrow: some believe that a further increase of concentration could strengthen national banks' market power and become a major obstacle to the full achievement of the single market (Cepr (1999)). There are many arguments against this conclusion. From a genera1 and theoretica1 point of view, the traditional negative relationship between concentration and competition (hinged on a simple s-c-p-model) can be reversed. When economies of scale exist and only big agents can implement effective competitive strategies, the relationship can be positive. More concentration can therefore lead to (even be the condition of) more competition (Ciocca (1998)). Although a systematic analysis of European markets is still lacking, al1 national surveys show a significant increase of al1 indicators of competition, accompanied by a dramatic decrease in the number of banks. In the years 199097 in the five big European countries, 2400 banks have disappeared (-26%). To explain this result one must remember that the previous decades of heavy structural regulation and public ownership of banks had created a very fragmented banking system in most European countries. In many cases (Italy being a notable example) the banking system was polarised among a myriad of local banks dominant (and profitable) in each market and a handful of big banks which could not count on a really national distribution network and had to concen-
trate on the more competitive segments of the market. Big banks were therefore very weak in the retail (and more profitable) markets. The merger wave in each European banking market must be seen as the natura1 reaction to the previous excess of regulation and as a major effort of opening up the market for corporate control in banking, as in the other sectors of the economy. Only through mergers banks have been able to correct the distortions created by regulation which dwarfed competition for so many years. On these grounds, it is difficult to see in the present merger wave only a "chauvinistic support for national champions" (Cepr (1999)). On the contrary, especially in countries which paid a heavy tribute to structural controls, most mergers must be seen as a defensive strategy, aimed to build an efficient distribution network and technica1 know how in innovation services, where competition is announced to be particularly strong. Further national mergers could be necessary to narrow the historical differences of main European banking systems. Also in this case, Italy is a leading example: while in the UK, France and Germany the market share of the top five banks is around 10-13 percent, in Italy it is only 5.3 percent (Banca d'Italia (1999)). In other words, structural differences among European banks are only the mirror of the distance from the achievement of the single market. What does that mean from a regulatory point of view? First of all, we do not have as yet a real European-wide regulatory problem, because we still lack real European players and real market integration. Both from the point of view of stability and protection of investors, problems are essentially domestic in scope. On the other hand, one can assume that the monetary union wil1 accelerate the integration: the second stage, characterised by large cross-border flows and even cross border tnergers has already begun. It is therefore urgent to wonder whether the present regulatory arrangement (fragmentation of regulatory authorities cum the home country principle) is suited to cope with the foreseeable evolution. Second, the completion of the first stage, i,e, the consolidation of national financial systems rests on the shoulders of national agents and national regulatory authorities. From this point of view, there are many evidences that the coming integration could enhance the protectionist attitude of regulators and the danger of regulatory capture. In other words, the higher the fragmentation of regulatory solutions, the higher the probability of a slowing down of the integration pro-
cess due to regulatory asymmetries andlor the intentional or unintentional propensity of individual regulators to protect national agents. This worry is clearly expressed in a recent paper (Cepr (1999)) aiid is shared by the Commission (1999) and soine influential members (Monti (1999)). These two observations together lead to the conclusion that the coming integration of European financial markets not only require to pursue the fundamental objectives of financial regulation in a European dimension but als0 raise the urgent need of harmonising regulatory structures and regulatory policies. V. A REGULATORY FRAMEWORK FOR THE SINGLE MARKET
The previous analysis has shown how important is to question whether the present regulatory structure is well-suited for the foreseeable evolution of the European financial system. The philosophy of mutual recognition not only lead to a limited convergence of regulatory framework in each country, but als0 increased the level of complexity, superimposing European bodies to the national ones. Adopting our classification of objectives we can identi&: 1) Price stability: in the 11 countries members of the Monetary Union it is the full responsibility of the European System of Centra1 Banks, according to the provisions of Maastricht Treaty. 2 ) Financial stability: rests on the shoulders of different national supervisors. This solution follows the principle of subsidiarity reaffirmed by the Maastricht Treaty.
The Treaty envisions only an advisory and consultative function for the ECB in the field of prudential supervision5. Generally speaking, rules aiming at limiting the role of ECB in prudential supervision prevail on any other rule. Looking at the situation of individual member states, one can say that only a few countries have changed their regulatory structure to adapt it to the evolution of the financial system, particularly as far as the objective of financial stability is concerned. As shown in Section IV, the variety of solutions in terms of responsibility, powers, modus operandi etc. has somehow increased. The examples of reforms aim-
ing at dividing responsibilities along the "horizontal" lines of objectives are still an exception. A "vertical" line, according to subject~,still prevails. In a few cases (Scandinavian countries and the UK), stability and disclosure responsibilities are assigned to the Same authority. The variety of solutions regarding supervision of insurance companies is still greater than in banking. 3) Disclosure: the situation does not differ significantlyfrom the supervision of financial stability. Unlike financial stability regulation, the question of if and how to establish a European authority has never been discussed. 4)
Competitiveness. There are two levels of competence. The European (for cross-border transactions) and the national one. Italy is the only European country assigning this responsibility to the centra1 bank.
The genera1principle to follow in case of services offered across borders is, as already mentioned, the home country responsibility. This picture contains al1 the relevant problems that are now under discussion at the European and national level. In particular, we can identifj: i) as to the objective of price stability: the optima1 allocation of lending of last resort functions; ii) as to the objectives of financial stability: the redistribution of responsibilities from the national to the European level; iii) as to the objectives of disclosure: idem; iv) as to the objectives of stability and disclosure: the suitability of the home country principles to the monetary union scenario. A. The objectives ofprice stabilig and lending of last resort Lending of last resort is the key component of centra1 banking. As Hicks remarked, centra1banks were born and developed as typical national institutions (and embodied the supreme manifestation of sovereignity: printing money)6. In a new supra-national environment how should lending of last resort be dealt with? Two approaches emerge from past, mainly domestic, experience, which stem from different theoretica1 views of the nature of bank crises, as the main unresolved dispute regards the defi-
nition of a crisis, not itsproper handling (Giannini (1998)). While monetarist~stick to a narrow definition, economists stressing uncertainty and the built-in danger of financial crises, suggest a wider approach, where lending of last resort should be granted in case of a great variety of disturbances such as a sharp decline in asset prices, the failure of a large intermediary, or a disruption in the payment system. In the first case, lending of last resort can (or even should?) be separated from supervisory powers. In the latter, it should not. As already mentioned, Germany adopted the separation approach and consequently the Maastricht Treaty adopts a "narrow centra1 banking" model. The two theoretica1 approaches are mirrored in two different judgements on this model. The first finds it inefficient and dangerous, on the following line of reasoning (Prati-Schinasi (1998)): a) the probability of liquidity and solvency problems is greatly enhanced with monetary unification and the single market because of: a l ) possible misfunctionings of the new payment system; a2) difficulties of banks put under pressure by an increase in competition. b) regulators will have to distinguish rapidly between liquidity and solvency problems within pan-European financial institutions and across European-wide financial markets. c)
this is not the most likely outcome, for three reasons, one specific to EMU, the third more genera1 cl) practices among centra1 bankers are very different; c2) the border line between the responsibilities of national centra1 banks and the ECB is not wel1 defined. c3) in particular it will be difficult to put int0 practice the traditional Bagehot prescription (lend freely, at penalty rates, to illiquid but solvent banks).
d) as a consequence, to avert the crisis and to follow Bagehot's principle, the lender of last resort must be better informed than the market. In the case of Europe, this better information should
come from a coordination between national supervisors, national centra1 banks and the ECB. The probability of a failure in such a coordination is high, to say the least. In particular, as Prati-Schinasi ((1998) p.35) points out: "the ECB appears to have either to make a decision about whether national centra1 banks should be allowed to intervene in a local liquidity crisis or to inject extra funds int0 the system in the event of a genera1 liquidity crisis". This requires access to information on the financial condition of the counterparty institutions. Supervisory information would be necessary to assess the credit risk that such operations would involve in the event that non-eligible collateral needed to be accepted. The ECB would not be able to rely on market assessment to distinguish between a liquidity and a solvency crisis. Therefore "the current information sharing arrangements between national supervisors and the ECB appear inadequate" (ibidem, emphasis added). Two solutions have therefore been suggested: i) to allow the ECB evolve (the sooner the better) into an institution that would assure a leading and coordinating role in crisis management overcoming the "narrow" concept of centra1 banking envisioned in the Maastricht Treaty (Prati-Schinasi (1998)); ii) to build up a real European supervisory body (CEPR (1999)). The second school of thought is wel1 synthetized in a recent paper given by a member of the Executive Board of the ECB (Padoa-Schioppa (1999)) and follows this line of reasoning: a)
stability problems must not be underestimated, but they do not require to exert jointly lending of last resort and supervision, because: a l ) Bagehot's time are over and "nowadays runs may occur mainly in textbooks", thanks to the instruments we have reviewed in Section 111; a2) the Eurosystem has developed a substantial capacity to react, under the coordination of ECB and national centra1banks; a3) banks have enough collateral to tackle dry-ups of liquidity deriving from a gridlock in the payment system or a sudden drop in financial market prices.
b) therefore "there is no expectation that the division of responsibilities between the centra1 banks and the banking supervisory functions should be abandoned".
The strongest point of this position is that Bagehot times are over and new instruments have been worked out both to control ex ante the riskiness of bank portfolios and to minimise ex post the loss for depositors. Although this view seems to reverse opinions formerly expressed by the present ECB's Chairman7 it seems to suit both the actual arrangements on the functioning of lending of last resort and the present distribution of centra1 bank responsibilities between the ECB and national central banks. First of all, one must remember that Europe is a regio11 composed of highly developed countries, where banks are fundamentally robust and the disclosure of banks' conditions and risks have been greatly improved. In a scenario like this, when a bank crisis occurs, it is less and less likely that supervisors have better information than the market. When the market is caught by surprise, the regulators are surprised as well. Emergency lending of last resort is activated on the basis of the judgement of the potential domino effects of the crisis, not of the internal situation of the ailing intermediary, as it has been demonstrated in the LTCM case. It is therefore compatible with the present "federal" architecture of centra1 banking that lending of last resort is distributed in the System and given mainly to national centra1 banks. But if we can dismiss the problem of lending of last resort at the European level and therefore conclude that the present arrangement should not be changed, at least in the short term, the problem still remains in the field of stability regulation. The strongest point of PadoaSchioppa reasoning is that coordination of centra1 banks in the lending of last resort function is wel1 rooted in past experience, can act very swiftly and efficiently, hinges on homogenous powers and traditions. This is the polar opposite of what happens in the other fields of regulation, namely stability regulation and particularly disclosure regulation. It is in these fields and particularly in the latter that we can find the most urgent need of European solutions.
B. Stability regulation: prudential controls In a global financial system, stability has completely lost the national dimension it had historically. This is why international institutions (BIS, IMF, the Basle Commitee) have seen their role reinforced and new coordination efforts have been made. In particular, through the Forum for Financial Stability promoted by the industrialised countries. Moreover, Iosco and the Basle Committee started a fundamental project for a proper supervision of financial conglomerates8. National regulators have strengthened their instruments, normally immediately after the emrgence (and solution) of a major crisis. As for the position of the European countries, two points must be made. First, while the global dimension is the only way to cope with systemic instability in a global financial system, in none of these international arenas, a common European position is represented. This is the evidence of the lacking of a European policy on supervisory matters which explains how far national positions prevail on a possible common European view. Second, the architecture of stability regulation still widely differs, thus creating a powerful obstacle to the integration of financial markets. First of all, only a few countries have adopted a distribution of responsibilities according to the objectives of regulation, as we have seen in the previous Section. Moreover, the modus operandi of nation regulators is even more different, leading to different regulatory costs, different regulatory cultures and dangers of regulatory capture. Finally, the regulation on specific segments of intermediation (mutual funds, insurance, in particular) is even less homogeneous than in banking (European Commission (1998)). In this field the process of harmonisation through the directives has been only partial (better: could not keep the pace with the market). Structural controls still prevail on prudential control. To sum up, European financial services are not fully harmonised, even in segments (such as asset management) which represent a growing proportion of the households' portfolio. European regulators follow different patterns, more inward-looking than orientated towards European integration. It is time for the European Commission to tackle the problem of stability regulation at the European level. The analysis has shown that the problem is not only to prevent systemic instability, as the recent Cepr report seems to imply. If this were the case, we could be happy
at least in the medium term with the present arrangements between the ECB and the national centra1 banks. In the wider field of stability regulation we can find a more urgent need for European solutions. The time seems to be arrived to have a supranational body (along the lines followed in the transition from the European Monetary Institute to the ECB. For the reasons shown in the previous Section, this body should be separated from the ECB. The first step could be to give a forma1 status to the Committee of European regulators in charge of stability, to endow it with a staff and a full-time Secretary Genera1 (not belonging to any of the national regulators) and to commit it to dealing with the problems of financial regulation in the European dimension. The new-born Committee should have a threefold mission: to bring a European voice in the international arena; to foster the legislative initiative at the European level; to identify the obstacles to integration, promoting as the case may be the necessary legislative initiatives. A similar solution could be adopted in the field of the regulation aiming at disclosure, which wil1 be reviewed in the next Section. C. Disclosure regulation Regulation aiming at disclosure raises different questions because of basic differences regarding the modus operandi of regulation. Controls aiming at stability must preserve investors' confidence on the financial intermediaries. They have to be used ex ante and based mostly on discretion and fiduciary information between the regulator and the regulatee. There are seldom "announcement effects" in regulation aiming at stability. When a rescue operation is necessary, the announcement arrives only when the "lifeboat" is safely in harbour. On the contrary, controls aiming at disclosure are fundamentally based on two pillars: ex ante, assuring a clear set of rules on the relationship between the intermediaries and their clients; ex post, assuring a real protection against alleged damages. The latter require the full publicity of interventions and is crucially dependent on things such as: speed of action; visibility of action; visibility of enforcement The European situation up to now leaves much to be desired, from both the ex ante and ex post point of views. As to tke former, the mutual recognition principle led the Commission to refrain from intervening in the content of national rules of conduct relating to the min-
imum information to be provided and the legal details of the finaiicial service (often built-in in the contract). The Commission does not seem ready to change this attitude, on the ground that harmonising rules of conduct would result in standardising financial products, which "would impair product innovation and be demonstrably out of touch with market development" (Financial Services Policy Group (1999)). This position is not theoretically robust: rules of conduct are meant to avoid information asymmetries between the intermediaries and their clients and do not bear on the characteristics of the financial products. As a matter of fact, they left enough leeway to the intermediaries to allow for financial innovation at the national level. One could wonder why a harmonisation at the European level should impair international financial innovation. If it is true that some ways of designing regulation in this field could end up in restraining investment firms' decisions on the conditions of the contract or the organisation of the firm, it is als0 true that most regulations proved to be "lean" enough to protect investors without damaging competitive forces. Putting it less mildly, this position proves that the position of the Commission seems to take int0 account more the needs of the providers of financial services than those of the investors. The consequence of the present Commission's attitude is to allow each intermediary to adopt its rules to market its services in other member States, i.e. to adopt a wider approach to the concept of home country control. As we wil1 show below, this is not a satisfactory approach to the ex ante balance of disclosure regulation between the efficiency of competition and protection of investors. As far as the ex post efficiency of disclosure regulation is concerned, the need for a European approach to disclosure regulation comes from the increase in cross-border transactions we expect to see in the near future. The linkages among Stock Exchanges, the desire of financial iiistitutions to enter new European markets, the demand for international diversification by European households, make only too easy to forecast a rapid increase of integration. Also in this field, we can find the asymmetries in the architecture of regulation and powers attributed to national authorities we have seen for stability regulation. There is a further problem, which refers to the necessity of coordination (and fast coordination) the disclosure regulation requires. The instrument so far designed to coordi-
nate the action of national authorities is the MOU (memorandum of understanding). The question therefore is: are MOUs an efficient way of tackling the day-by-day problems of intermediaries supervision at a European level? If the answer were yes, one should als0 think that barter is more efficient than money. Which is the polar opposite of what our textbooks say (and in this case they are still right). In fact: e
e e
the working out of the "understanding" is a very time-consuming process. Memorandums are full-fledged treaties, with al1 the diplomatic paraphernalia so they cannot be changed as promptly as the market evolution requires. in a n-countries area they imply n*(n-1)/2 bi-lateral relationships; the idea is that tlie host country should serve as the longa manus of the home country. But the action can find a fundamental weakness in the difference of powers concerning the conduct of the enquire.
MOUs are therefore a second-best solution. If this can be acceptable in many fields of international relationships, it is much less acceptable in a monetary union, whose goal is to achieve a single unified market for financial services. Europe therefore needs a coordinated action in disclosure supervision and this coordination can be inspired only by a fully recognized institution. A first step has already been made in 1997 when a European Forum (Fesco) was created building on the experience of a past (informal) committee of securities regulators from the European Union countries. The Fesco should evolve in an autonomous Committee, along the lines discussed for stability regulation. Its mission should be the Same as the stability regulation committee and the handling of cross-border litigation between intermediaries and their clients. The remaining question is whether this Committee should be separated or not from the Committee for stability regulation. The answer depends on the optima1 solution one wishes to give to the regulatory architecture: a two-headed or a single regulator? As we have seen the former model seemed to prevail until a few years ago, but it has not been fully implemented. The latter has been adopted only by the UK and Scandinavian countries. It seems therefore more sensible to give birth to two different committees, one for stability regulation, one for disclosure regulation.
There should be frequent and thourough exchange of information between the two committees. A merger int0 a single body should eventually come if and when a "single" regulator model wil1 become the rule instead of the exception. This new agency could als0 find a workable solution to the big battle on the host country principle which wil1 probably be one of the first points on the agenda of the next Commission.
D. Whose rules? The tangled web of home country control As we have already seen, the principle of home country control is the cornerstone of supervision on an international basis where two or more authorities are involved. The problem is that the principle works reasonably wel1 between banks of different countries. It does not deliver the Same results in the case of financial services within a single market. What does home country responsibility mean when disclosure and rules of conduct regulation apply? Whose "home" must prevail: investors' or intermediaries'? The first answer is that, because this regulation is meant to protect investors, it is the regulation of the countrywhere the service is provided, i.e. the country where demand meets SUPP~Y
But the reality is a little bit more complicated than that. We have four possibilities:
e
e
the service is offered by the local subsidiary of a foreign intermediary (whether investment firm or bank) incorporated in the host country; the service is provided by a foreign intermediary operating through a sales network incorporated in the host country; the service is provided by local workers (tied agents) of the foreign investment firm; the service is provided over the phone or via internet.
Only in the first case it is easy to identify a clear-cut distinction between home country and host country rules. In al1 other cases, possible conflicts arise. In short, the dilemma hovers between two equally paradoxical and unsatisfactory solutions.
e
either the investment firm is forced to comply with 15 disclosure regulations if it wants to offer investment services throughout the Union; or the investor must compare 15 disclosure regulations in order to determine which one is best suited to his or her needs of protection ex ante and ex post.
The trade-off between fundamental goals of regulation is therefore striking: competition on the one hand; protection of investors on the other. A few regulators, namely the British, are adamant about the possibility (opportunity? need?) to adopt the regulation of the country where the service is providedg. This position is substantially backed by the European Commission, which - worried by the danger that individual member states use local rules to protect local firms, seems favourable to allow an extensive use of the rules of the provider of the financial serviceslO.Although the worry of a protectionist attitude by individual regulators has found many evidences also in the recent past, one must not forget that - as the analysis above has shown - a home country control stricto sensu can be contradictory with the fundamental goal of protecting investors. The response to this problem requires a further effort of harmonising European regulation in the field of disclosure and rules of conduct. This is the major task that European regulators are to face in the years to come. This response could come from ad hoc European regulatory bodies which have been reviewed in this Section. From the institutional point of view this is, to my mind, the most important and urgent task. We do not need to give the ECB supervisory powers. But we need to coordinate stability and disclosure regulation and to begin to build regulators with powers and responsibilities suited for the Single Market for financial services. NOTES 1. In the classificalion suggesied by Llewellyn (1999) spccial ernphasis is given only to thc objectives b and c of our scheme. According to Llewellyn (p.10): "two generic types 01 regulation and supervision are identified: I) prudential regulation which focuses o11the solvency and safety and soundness of finailcial institutions and ii) conduct of business rcgulatiori which focuses o n how financial firms conduct business witli their customers". 2. "Conduct of business regulation and supervision focuses upon how financial conduct business with their customers. If focuses upoii mandatory information disclosure, thc honesty and integrity of firms and their employees, the level of competence of firnis
3.
4. S.
6.
7.
8.
9.
supplying fiiiancial firins and products, fair business practices, the way financial produ c t ~are marketed, etc., . Overall, conduct of business regulation establish guidelines aboiit appropriate beliaviour and business practices in dcaling witli customers" (Lllewellyii (1999) p.11). Wymleersch (1998) p.30-31: "11 appearecl to legislators in several states tliat prudential and transactional supervision could iiot remain domiciled in the hands of the sanie supervisor~body .... This separation of functions also avoded the risk of a biased prudential approach, flowiilg froin the prudcntial supervisors dictating tlie terms of the transaction. Integration of priidential supervision in the hands of the banking supervisors can further be jiistificd on tlie grourids of cfficiency, expertise and possibly also greater reliability.., In most systems, prudential supervision has been put in tlie hands of the prudential supervisor of the credit institution. Prade (1997). He was echoed by the Rapport de la Martinière: p.12 "Le moment paraît venue de relacher lc contrôle des produits pour reporter l'attention de l'autorité publique sur Ie contrôle des acteurs". In particular, art.3b states that "in areas which do not fa11 within its exclusive competence, the Community shal take action only if and so far as the objectives of the proposed action cannot be sufficienlly achieved by the Mcmber States and can therefore, by reason of the scale 01- effects of tlie proposed action, be better achieved by tlie Community. Moreover, art.25(1) envisions a specific advisory function for the ECB in the field of Community legislation relating to tlie prudential supervision of credit institutions and the stability of the financial system. Art.105(4) contemplates a somcwhat stroliger role for the ECB by stipulating that it must be consulted on draft Community and national legislation falling within its field of competence. Art.105(5) makes clear that the role of the ESCB is subordinate to that of the competent supervisory authorities by indicating that the ESCB is excpected "to contribute to the smooth conduct of policies pursued by the competent authorities relating to the prudential supervision of credit institutions and the stability of the financial system". Art.195(6) limits the role of the ECB in the area of prudential supervision to "specific tasks" tliat the EU Council may confer to it on a proposal of the European Commission. "Central bank suffer from the limitationh that they are national central banks. Only in a national econoiny that is largely selcontained, can a central bank be a true central bank; with the development of world markets, and 'especially) of world financial markets, national centra1 banks take a step down, becoming single banks iil a world-wide systein, not at the 'centre' any longer. Thus, the problem that was (partially) solved hy the insiitution of national centra1 banks has reappeared, and is still unresolved (though we are trying to solve it) on the world level." (Hicks (1967)). Duisemberg (1995), cit in Prati-Schinasi (1998). "In my opinion, banking supervision is a central bank function. The conibinatioii, within the central bank of banking supervision with lender of last resort, oversight and monetary policy functions offer distinct advantages. These advantages should not be ignored, considering the significance of financial stability - especially within an open and libcralized economy - and the coniribution wliich banking supcrvision makes in tliis respect." "The emergence of corporate groups which provide a wide range of finaiicial services, known as financial conglomerates and typically incorporating at least two of banking, securities and insurance, has created an additional dimension for the solo supervisors of entities within those groups" (Basle Committee - Iosco (1999) p.6). Davies (1998): "Where conduct of business in concerned, the rules which apply are those of the jurisdiction where the service is provided. However, as yet the EU has not defined how to decide wliere a service is provided so there is no EU-wide consistency. Because most business is still done domestically, where a firm is doing business from a branch this means that generally the rules of tlie state regulating the branch will apply. They will also tend to be ihose of the state of the consumer. Also, the supervisor there wil1 have a direct relationship with the branch so it wil1 he wel1 placed to enforce those rules. However, this structure is mucli more problematic where the financial institu-
tion concerned provides a service into another jurisdictioii withou a physical preseiice there. In tliat case, deciding whose rules apply - a crucially enforcing tliern effectively against a firm - is more difficult. In its banking coinmunication, the Cornmission lias suggested that the rules which should apply sliould be those of the state where thc business is actually done. We would agree with that." 10. Financial Services Policy Group (1999): "While the underlying assumption must reinain that fiiiancial products lawfully markctcd in one Meinber State sliould be freely available in al1 (tliis flows from tlie basic principles of the Treaty and from the concept of freedoin of services inparticular) tliis freedom can be liiiiited by local rules which foreigil service providcrs must follow as long as they are justified by e geiieral good. I-lowever, it is not an uiiqualified right for Member States to insist on uilwavering compliance with local provisions on conditions, content and marketing of financial services. Jurisprodencc of the ECJ clearly estiblislies that local rules niust not provide disporportionate protcction of the general good and must take iiito account analogous rules that are applicable in the country of origiii and which protect tlie same interest (mut~italrecognition)."
REFERENCES Abraham, Jean-Paul, 1999, Le strategie delle banche in Eurolandia, Bunca Impresn Societri, 1. Banca d'ltalia, 1999, Relazione Annuale, Considerazioni finali del Governatore, (Roma). Briault, Clive, 1999, The Rationale for a Single Financial Services Regulator, FSA Occasional Paper Series, 2, (London). Bruzzone, Ginevra and Polo, Michele, 1998, Sources of Market Power in Banking: Tlieory and Implications for Antitrust Analysis, (Ente Einaudi, Roma), mirneo. Centre for European Policy Research, 1999, The Future of European Banking, (Cepr, London). Ciocca, Pierluigi, 1998, Coiicorrenza e concentrazione nel sistema finanziario italiano, Banca d'ltnlia Bollettino Economico, October. Commissione Europea, 1999, Servizi finanziari. Messa in ato del quadro di azione per i servizi finanziari: piano d'azione, (Comunicazione della Commissione, Bruxelles), COM(1999) 232, 11.05.99. Consob, 1998, Relazione Annuale, (Milano). Davies, Howard, 1998, Introduction in Fiiiancial Services Regulation: a Europeail Perspective. Thc Financial Services Authority Conference, (London, SFA). Dewatripont, Mathias and Tirole, Jean, 1993, The Prudential Regulation of Banks, (The MIT Press, Cambridge). European Commissioii, 1998, Financial Services: Building a Framework for Action, (Bruxelles). 1997, Financial Services Autliority: an Outline, (London, FSA). Financial Services Policy Group, 1999, Retail Financial Services: Remaining Barriers, Working Document n.6, (Bruxelles), mimeo. Ghezzi, Federico and Magnani, Paola, L'applicazione della disciplina antitrust comunitaria al settore bancario, (Ente Einaudi, Roina), rnimeo. Giannini, Curzio, 1998, Enemy of None but a Coinmon Friend of All? An International Perspective on the Lender-of-Last-Resort Function, (Banca d'Italia, Temi di Discussione del Servizio Studi, Rome). Goodhart, Charles, 1985, The Evolution of Ceiitral Banks, (London School of Economics and Political Science). Hicks, John, 1967, The Two Triads, in Collected Essays in Monetary Theory, (Clarendon Press, Oxford). Llewellyn, David, 1999, The Economic Rationale for Financial Regulation, (FSA Occasional Paper, London), April.
HM Treasury, 1998,Financial Services and Markets Bill; a Coiisultation Document, (HMSO, London), July. Monti, M., 1998, The Future for Financial Services in Europe: Challenges for Markets and Regulation, in: Financial Services Regulation: a Europeail Perspective. The Financial Services Authority Conference, (SFA, London). OECD, 1998, Enhancing the Role of Competition in the Regulation of Banks, (OECD, Paris). Onado, Marco, 1992, Economia dei sistemi finanziari, (I1 Mulino, Bologna). Prati, Alessandro and Scliiiiasi, Garry, 1998, Financial Iristability in EMU, mimeo, (Università Bocconi). Padoa-Scliioppa, Tommaso, 1999, EMU and Banking Supervision, Lecture at the London School of Economics, 24 February 1999, (ECB), mimeo. Prada, M., 1997, Regulatory Responses to the Integration of Financial Services, Iosco Annual Conference, (Taipei). Taylor, M., 1997, Wallis Commission Takes the Twin Peaks Route, Tlze Finn~zcinlRegillator.2, 1. Wymeersch, E,, 1998, The Implementation of the ISD and CAD in Natioiial Legal Systems, in Ferrarini, G., ed, European Securities Markets. The Investment Services Direcitve and Beyond, (Kluwer Law International, London).
Tijdschrift voor Economie ei1 Management Vol. XLIV, 3, 1999
Competition and Cooperation Among European Exchanges The Impact of Pinancial Globalisation with Special Referente to Smaller-Sized Exchanges J.-P. ABRAHAM" and A.-E PIRARD**
In Europe, the traditional image of a stock exchange as the institutionalised meeting place of traders, operating on a sometimes frantically agitated trading floor in a temple-like building, is fading away at a very fast pace. Electronics replaces open-outcry; deregulation erodes the public status and the privileges of an institution, the regulations of which, in some Continental Exchanges, date back to the Napoleonic era. Quote-driven dealer markets like EASDAQ challenge the order-driven auctions of the official stock exchanges in terms of easy access and flexibility. Electronic networks (ECN) even undermine the very concept of a physically located organisation. Structural change is on hand, the euro being the catalyst by creating a (potentially) large market without exchange risks and costs.
"
-
Departement Economische Wetenschappen, K.U.Leuven en Facultés Uiliversitaires Nortre-Dame de la Paix, Narnur. *" Studies Department Artesia Bank, Brussel. Wc are very grateful to Hans Brouwer, Francesco Cesarini, Marc Corluy, Martin Fase, Hugo Lasat, Olivier Lefebvre, Frank Lierman, Michel Maquil, Jean Peterbroeck, Anne-Sophie Pycke, Vincenl Van Dessel and Ann Vlemincla for the contacts they organised, tlie comments they gave andlor tlie information aiid statistics they l
No wonder that, under such conditions, exchanges, while still performing some public duties, become firms and act as such; like other financial firms, they compete, cooperate and sometimes even merge. Hence, their behaviour and their development can be analysed, at least partially, on the basis of the micro- and meso-economic theory of firms and markets. Part I1 of the present essay is mainly devoted to such an analytica1 exercise on the change in the micro-and meso-structure of the European stock exchanges. It shows how some technological, socio-economic and institutional driving forces interact in a quasi-dialectic process, to extend financial globalisation, increase contestability, foster competition and ultimately force exchanges to adjust and to change. Part I11 is more empirical. It discusses the reactions of European exchanges to the challenge of structural change. It concentrates on the moves towards setting up a European integrated system of exchanges, either by interconnection or by creating a pan-European platform. Our special attention goes to the future of the smaller-sized exchanges. We claim that, in order to ultimately suivive in a struggle, in which too many exchanges chase too few representative stocks, smaller exchanges have to intensify and institutionalise their cooperation. In other words, strenghtening a common profile becomes a must. 11. DRIVING FORCES AND FINANCIAL GLOBALISATION: THEIR IMPACT ON EUROPEAN STOCK EXCHANGES In this part of the paper, we analyse the basic features, the relative strength and the interaction among the driving forces, which we consider as the most important under present conditions. Aftenvards, we link the impact of these forces with the process of financial globalisation, and derive some implications for the various types of exchanges in Europe.
A. The main driving forces in the financial services industry It is widely accepted that, at the present moment, a set of strong interactive forces exert a determinant influence on the developments in the financial services industry. Directly or indirectly, they are shaping the future of the European stock exchanges (Exhibit 1).
EXHIBIT 1 Dri1,ingforces ofstnictural change in Europea~zstock exchanges
7. Regulatory developmenls 3. Euro(pean) integralion 4. Specific struclural factors in the financial services indiistry, related to basic socioeconomic trends as tbe ageing of the population, the reduction of traditional intermediation by banks, lhe spread of an equity culturc in continental Europe, etc.
Progress in information and telecommunication technology has tremendously extended the possibilities of registering, handling and transmitting information, data, trading proposals and agreements, at a very low cost. This statement obviously applies strongly to exchange-trading. Early this year, the press (Wal1 Street Journal (1999)) reported that the recently installed SAX-2000 trading system at the Stockholm Stock Exchange was theoretically able to absorb 2000 registrations per second, compared to only 10 in the former SAX-system and 110 in the more advanced German XETRA. The cost reduction induced by electronic trading is substantial. The cost of electronic screen-trading amounts only to 40% of open-outcry cost. Electronic Communication Networks are operating at a cost of maximum one tenth of organised auction systems. Even more important, network economists emphasise that (electronic) networks operate under conditions of positive size externalities, which, under competition, stimulates their extension. But technological gains can been fully reaped, only in a non protectionist, deregulated environment. The technica1 properties of SAX2000 could not be profitably exploited if the Stockholm Exchange had remained a segmented and protected corporative institution. It is no coincidence that the SAX-2000 has been set up in a demutualised organisation, which is even listed on the stock exchange and has recently announced a gradual (implicit) merger with the Copenhagen Exchange in NOREX. Regulatory developments, our second driving force, are therefore of overriding importance. Eurobeun) integration is often analysed as an important factor of external deregulation. The Single Act (1986), the Maastricht Treaty (1992) and the CAD and ISD Directives (1992-1996) have indeed removed the last obstacles to free capita1 movements in the EU. Through mutual recognition and the European passport, they also introduced the basic building blocks for a single market of financial services. Last
but not least, the introduction of the euro transforms this single market into a financial market without exchange risks and costs. Finally, some structural factors more specifically concern the financial services industry. Thegreying of the European population stimulates savings through mutual or pension funds. Institutional investors have become important financial agents who, as recently suggested by Allen and Santomero (1998) deserve more attention in the theory of financial intermediation, where the role of banks is diminishing. Equally important for the stock exchanges is the spread of the socalled equity culture. Obviously, Euroland has to catch up with respect to the Anglo-Saxon world. In 1997, the market value of equities in the former only amounted to 18% of the total financial system, compared to 39 % in the USA. (Artesia (1999)). But change is on hand, under the pressure of two interactive and convergent factors: more demand for investment in equities from institutional and private investors in periods of low inflation and low interest rates, and more willingness of enterprises to finance their expansion through the stock exchange rather than through the banking system (Artesia (1999)). In Exhibit 1, the order of enumeration reflects our subjective evaluation of the relative strength of the four driving forces, in the actual situation. This relative importance is not permanent but changes, sometimes abruptly, over time. Major institutional breakthroughs, such as the Europe 1992 project implementing the Single Act, or the introduction of the euro are discontinuous by definition. They capture the attention and dominate the strategies of financial agents at the time of the breakthrough but they sometimes fail to keep this impact over time. We believe, however, that the euro-effect wil1 have a more lasting impact than the single market-effect (Abraham (1999a)). In the Nineties, technological progress bas not only been fust, but it has also become more and more endogenous and even continuous. Technology explains the origin and the development of dealer exchanges like NASDAQ and EASDAQ, full electronic trading through ECNs, as wel1 as the successful adjustment of traditional order-driven continental exchanges in their competition with SEAQ International. Technology is omnipresent in the securities industry, even more than in banking: virtual, fully electronic, exchanges, without precise physical location, seem to be more near-by than virtual banking, as shown by the willingness of the American Securities and Exchange Commission to recognise, under some conditions, ECNs as full-
fledged exchanges. Finally, specific structural change linked to demographic or socio-economic developments is also continuous but slow. This analysis induces US to conceptualise the interplay of the various driving forces in the financial services industry as a qzlasi-dialectic process, which reminds US of the regulatory dialectic, popularised for banking by Prof. E.J. Kane (1981) and his followers at the beginning of the Eighties. Two key features can be distinguished in this process: (1)the interaction between technological and institutional forces with its impact on the behaviour of financial agents and (2) the combination or the trade-off among four aspects of exchange performance: cost, liquidity, immediacy and transparency. In a highly segmented and protected corporative exchange, transparency and costs are rather high (regulations, fixed commissions) wheras liquidity and immediacy are rather low (not much free float, intricate trading and settlement procedures). Technological progress is hampered by corporative resistance to innovation (open-outcry against screen-based trading) and lack of investment (no profits to be ploughed back in technology). When the protectionist, regulated corset is removed, markets become larger and deeper, but competition increases. Technological innovation becomes a strategic factor in the struggle for market share and profits. New participants enter the market of marketplaces: dealEXHIBIT 2a Operational efliciency factors in exchange-trading
er exchanges providing easy access and immediacy for large transactions, ECNs and other proprietary systems providing low cost but often less transparency and liquidity. The established exchanges which accept the challenge of competition become the champions of liquidity in its relation with volume, automation, transaction costs, transparency and transaction risk (Exhibit 2a). As the newcomers often operate at decreasing marginal cost, competition threatens to degenerate int0 a cut-throat struggle, the outcome of which tends towards a monopoly for the surviver. At this stage, the established exchanges, in order to meet the challenge of the newcorners, are forced to integrate quote-driven and pure electronic trading, either as a parallel or a complementary (e.g. after business hours) segment to their Premier Marché. This year, even the morher of al1stock exchanges, the NYSE, had to tackle this problem of joining, instead of beating, electronic trading. Meanwhile, on the demand side, the risk aversion of institutional and private investors has diminished. Risk capita1 Ilourishes and an equity culture develops. At the Same time volatility, overshooting and financial scandals multiply. Once more, stability has to be weighed against efficiency. At this stage, an institutional reversal may occur. Supervisory authorities face a dilemma: should, on the basis of the caveat emptorprinciple, market participants bear the full responsibility for the risks they have taken, or should priority be given to the protection of the smal1 investor, in a world where the equity culture extends to large and less sophisticated segments of the population? I[f the latter option prevails, the road is paved for a reregulation, which is not a return to protected and segmented markets, but which, even at higher cost, stresses transparency and integrity of markets, through more disclosure, more information, prevention and repression of fraudulous practices etc. (Onado (1999)) The wliole process is summarised in Exhibit 2b. This is the background against which the competition and cooperation problem of the European exchanges arises. Of course, our dialectic presentation reinains very sketchy. It should be completed and qualified in the light of the experiences in various countries and on various exchanges. The iiisights given by new theoretica1 developments should also be tested and integrated. Applying network economics to the stock exchange issues seems particularly promising under present conditions, where electronic networking is developing in al1 kinds of markets. Network economics, which originated in the eco-
EXHIBIT 2b Replntoly dialectic of European exchanges HISTOKICAI. I'ROFILE : MONOPOLISTIC INSTITUTION REGULATED ANI3 PROTECTED BY STATE ANDIOR CORPORATION
+
[,,,,,,,,,E= SIZE EFFECTS
I
I
I
INCREASED COMPETITION WITH EXTERNALITIES
MORE VOLATILITY FINANCIAL SCANDALS
L
I
I
l
REREGULATION : MORE TRANSPARENCY, MORE W E ? ' INrEGRiTY
I
nomic analysis of transport and telecommunications, stresses the importance of positive size extemalities: when the network enlarges, the possibility of al1 users to acquire more and better services from the network increases. Applying this principle to the financial networks, Economides (1993) distinguishes two kinds of externalities: (1) liquidily enhancement by size expansion, along the lines indicated in Exhibit 2a7 and (2) underpriced provision of market price information to outside rivals, which stimulates free riding by newcomers. According to Di Noia (1998), followed by IRS (1999), two implications emerge: (1) the importance to reach a minimum dimension in order to enjoy network externalities. Hence, liquidity tends to concentrate (Networksare by their nature self-inforcing Economides, op.cit., p. 4)) and (2) the impossibility to reach optimality in such a system of externalities: Network externalities rnay lock-in exchanges into insufficient outcomes, even in perjiect competition"8If no agreements are found, competition among stock exchanges rnay end up with a monopolist exchange (Di Noia (1998) op.cit., pp. 42-43).
The relevance of such a type of analysis for a European market, where too many exchanges chase too few representative stocks is fairly obvious. B. Financial globalisation in Europe. Challerzgesfor smaller-sized exchanges Tn this section, we analyse the various ways in which the interplay of the driving forces, which we described above, produces financial globalisation and we also discuss the challenges which the smaller-sized exchanges in Europe have to face in this respect. The traditional definition of financial globalisation stresses the internationalisation of capita1transactions and markets. It has been popularised, e.g. in several I.M.F. publications: "8Globalization refers to tlze growing ecorzomic interdependence of countries worldwide through the increasing volume and varieq of cross-border transactions in goods and services and of international capitaljlows, and also through the more rapid and widespread use of technology"8 (I.M.E (1997)p. 45). In the present essay we use a more extended definition of financial globalisation, which covers three aspects: e
o
intemationalisation of capital flows, as defined above; f~~nctional globalisation of financial services, e.g. by bundling int0 one organisation the cash market transactions, tbe derivatives and the clearing, settlement and custody services. This is exeinplified by the merger of these services in theAmsterdam Exchanges (AEX), in the Brussels Exchanges (B), in Paris-Bourse-SBF; globalisation through the single currency, the euro, which refers to the specific issues of Euroland.
These three aspects are interlinked, but nevertheless stress different relevant features for the European exchanges. What is at stake - in a specific European setting - is described as follows in the 1998 Capita1 Marltets Survey of the I.M.F.:
1. an increase in the technica1 capabilities for engaging in precision finance, that is for unbundling, repackaging, pricing and redistributing financial risks; 2. the integration of national financial markets, investor bases and borrowers into a global financial market place;
3. the bluwilzg of distinction between financial institutions and the activities and markets they engage in; 4. the ernergence of the global bank and the internationalfinancial conglomerate, each providing a mix offinancialproducts and services in a broad range of markets and countries (I.M.F. (1998), Annex V, p.180).
In Exhibit 3, we link the four driving forces, analysed in section II.A, with the three aspects of financial globalisation, just mentioned. Without entering int0 details - some of them have already been analysed we may point out some general features of the globalisatisation process, which result from this systematisation: 0
0
0
0
Technology on its own has no unequivocal impact on financial globalisation. It allows for fragmentation and dissemination (e.g. through proprietary sytems) as wel1 as for bundling and concentration (through economies of scale); In combination with other driving forces, technology has more distinctive effects. The combination of technologicalprogress and liberalisation of capital movements favours internationalisation of financial services which, in its turn, induces economies of scale and stimulates concentration. It may contribute to the delocalisation of institutions and of high mobility, less customer-linked services (Walter (1998)). Electronic networking does not only cut costs in general, but, by doing so, increases the number of high mobility services (e.g. clearing, settlement, custody); Deregulation and EUDirectives provide opportunities for functional globalisation through branching out of activities and, more in general, by permitting remote access to the other EU exchanges, which secures direct information, lower transaction costs, coìrelation with other exchanges, increased liquidiìy and betterprice-discovery (Di Noia, op.cit., p. 22); The introduction of the euro opens the prospect of trading without exchange risks and costs. However, its net impact on the activity of stock exchanges may be ambiguous. Mergers and acquisitions of industrial firms induce delistings of vested companies. In the other direction, the European ambitions of fast-growing enterprises direct them more and more frequently to the stock exchange.
EXHIBIT 3 Llriving forces, financial globalisaliorz nnd the challe~zgesfol- srnalk7r-sized exclzanges IMPACT O N FTNANCIAL (iLOIIAI.ISATION TIIKOU(;II
DIIIVINCi FORCE
INI'FI
iiiic\iif \calc aiid rictwoih exieinaliiic\ o11 a world\vide b;iaia - iicu- oppoiriinitiei lor di>cr\ifi~,itiuaiof ooritolio\ -
F U N C l'IONAL (;LOBALISATION -
-
I:IIR«
iiri uiiivocal cffccl on tunctional glohalisatioii: ti-chntilogy pruvidc\ iiplwriuilitics t o r iinhundIing a \ wcli ;is tor buridling i ~ ï - Iccliiiology liecomei s1r;iiïgic iuiictioiis arid reivice' (AEX. factor iiíeniiniiccd cumpetiiioii BXS veiiiii C E I I P I . aiid EUin a (i~iiirc)Iraiispareiit maiksi KOCLEAR) clcctroiiic tiadiiiu tlircaiciis
ilirougli rcductioiid eritiycciita - itimiilui io incicascd iiiteniatio-
CtIAI~I~I~N(i1~S P O R SMAI,LFli EXCHAN(ikS
- rliilinli~;iIiori r,ii\ci Llic i i i i i i i k o m s i i c io Ihc 'ioiiiciiody iii tlic 1>1url l l i i -
~p,rlihctw r e n Ii>wcrco5ls aiid les5 litiuicliiy aiid Iranap,ircncy
KECiULATOI<\r DEVLLOPMEN I S cai\iiiig dci:c;itiaikdiii>ii iiilpriidcii1i;il cirnlrol :ti tlic ii;iliiiiial level 11011 - dciiiiiirialisaiion a\ ;i c;iiisc niid a cuiircqoence of cíimpclitiuii - prcïciit ,?,ie#iiiiiroii Si>curcr uil i ~ i i c g i i i y«I m,irkets ihiougli disclosuiz. inriiiinatioii, prcvriiiiiiii :,lid rsliiabsi~ii<>i lrsiidtilut,s p1 L,Ci,'C\
DRTVING FOIICE
IMPACT O N 1:INANCIAL ULOBALISATION T H R O U G I I INTERNATiONALISATIO~ -
-
Single M n r k ï t Ac1 11986) aiid Maastricht l i e a t y (lOOl-1992) \timulated: . lul1 Iil>eralisatiun oTEU cani-
these I~ICLOIS vcry o f t ï n rcflecl iiiieriiatiunal, mutually reiniorcing trends (ageing of tlie population, shift trom H tri M financial sy\lcnis, spread u i tiic cquity culiiire, etc.)
Sui~rce:Ahrdhani (19'1'1), IRS (1999). Williamson (19~19).
FUNCTIONAL GLOIIALISATTON
EURO
straicgic importniicc oi dircciivcs (CAD atid ISD Ilireclives 1'1'121996). . basic pniicililes ut mutual r ï cil~nilion . home coiiiitrycoiilral and Eoropenn passpiiil tur Ilic siixk exchanges niid ihcirniemhers: idcntic.11 niiei lor credit instituiioni and investment services lirnis - cross-border opporti~iiiticitu hraiic!~out in prirtnei couiitrics - cioss-border rcniuic íclccironic) accsss tu stock cxch;ingcï «i partrier coiiiitriei - «iganisatiori of oii- aiid o[[-exchange t r a d c willi p i i c e disciosurc
- electr(iin)ic \tiick of euro introiiuctir~nas (; milestoiic iii t i i r a peari iiiinricial iiilcgralion: coii~o1idatioii:i2nd nicrvers rcducc
-
- piomotion of fiiiictioiinl glob;rIi>atioii liiroiigh: . iiicrcaacd iiunibei ol paiticipants (iiistiiiitional investors, ...l increased iiiimbcr of produclr (siiares. buiids, dcrivativcs, mutual lunds. insi~iance) . diversitication o1 iniïi-niïdiaricï. procedilres and netwurks (internet!)
-
C'I-IAI.1.ENGES F O R SMALLER LXCHANCrES
- Eiiro(pe:in) iiiiegr;itiiiii stroiiply coiili-ihiiles 10 coriccirlratc ir;iiliiig in seiuiitics in ihc cxcliaiia n d tlie fiii;iiici;il centra\
ve tliem tri Ihe stock c~chiinpe locnli~,itiuii»i ECB in Fraiiliiiirl s u p p « r l s lliil enIergeilcc o1 IIciii\che Borse ;is tlic tiirolica n "rising star"
- easy acces aiid !ranip;iioiicy o1 Euroland markets eiiliancc ihc impact ol 1Iic ciiliei h c t o i s
- challc~>gc ti>ilipport ; ~ n d!o &;innel Ilie ?predil o f llic eijoiiy cllllllr'! - prcssoi-e on aller excli:iiigc\. especially riie 13ciielii.: iehanges 10: siiengtiieii tlie liiiki willi in\titutic~i~al it~vcsiors monitor srn;illcr liriii\ low:~rd\ !he markct of 5ni:ill :ind rnidc.ips \nsci,iliic in pin
In spite o£these ambiguities, we suggest that a global picture emerges as the most likely olie. Techiiological progress has a clear size-effect through concentration of trading and processing in trading systems of increasing capacity and performance. Large opportunities for higher liquidity and lower costs still exist. Deregulation and European integration just provide the iiistitutional framework to seize these opportunities and to materialise these positive prospects. Soine specialists (e.g. Cappon (1999)) do not see the competitive edge, the lasting value of exchanges along the lines we have followed in the preceding reflection. In their view, the role of the exchanges as a locale for trading will diiniiiish in an all-electronic world. The expansion and even the survival of the exchanges in their competition with pure electronic trading will mainly depend on factors of quality and reputation, in the provision of: e a regulated environment, which normally offers transparency and so more guarantees of fairness and compliance; e a convenient locus for regulators to carry out their duties; a badge ofprofessionalism for certified members and mernbership; s efforts of product R & D, product design and standardisation; marketing and education services at the product level. While fully supporting this quest for quality as a distinctive feature of established exchanges, we consider this approach as insufficient. It misses a solid quantitative basis for the exchanges as ultirnate suppliers of liquidig. It also neglects the experience and the lessons of the early Nineties when the continental exchanges staged a formidable comeback (Pagano (1998)) against the London-based dealer market SEAQI, by introducing continuous electronic order-driven systems, by liberalising access to membership, by liberalising fees and reducing transaction taxes. This cross-border competition has been considered as the engine ofchange in European stockmarkets (Pagano (1998), p. 200). Even in medium-sized exchanges, like the Brussels one, the information about cross-listed Belgian equities Ilew from Brussels to London, with Brussels providing the final liquidity (Jacquillat, Gresse and Gillet (1998) and Degryse (1999)). We do not see why European exchanges would be unable to stage an analogous reaction to aggressive proprietary networks. This certainly applies to the main EU financial centres: London, Frankfurt, Paris and it applies even more, if these three centres merge an important part of their orders into one pan-European platform.
Not enjoying the saine size-effect, the smaller-sized exchanges find themselves in a more delicate position: e an important part of their most rewarding business may migrate to the Big Brothers or be diluted in electronic trading; c they will have to face an increased pressure on costs, fees and taxatioil, in order to remain competitive; e they will be forced to specialise on securities with a strong national or regional flavour: companies with local recognition, smal1 and midcaps; e In this way, financial globalisation transforms the competition among exchanges in a struggle for ultimate expansion and even survival. SIS. COMPETITION AND COOPERATION AMONG EUROPEAN EXCHANGES AT THE END O F THE NINETIES A. Setting the stage How does financial globalisation bear in practice on competition and cooperation under present conditions, at the end of the Nineties? In order to make the transition from an a priori to an empirica1 analysis, from abstract reasoning to facts, some genera1 characteristics of the present stock exchange picture in Europe are to be taken into account. We highlight four of them: e
(unequal) Spread of the equity culture
Equity culture penetrates everywhere but progresses froin quite different starting points and at unequal speed in the various countries. As shown by the Exhibits 4 A, B, C in the appendix, the ratio market value of stock market capitalisation/GDP is still much higher in the United IGngdom (155 %) and the Netherlands (130 %) than in most other on-shore continental countries (Germany 39 %, France 49 %). On the contrary, in the latter countries, the increase in the Nineties lias been faster. However, as stressed in the Artesia study, a large part of the recent increase in the ratio is due to a price-effect because stock exchange indices have risen much faster than the GDP price indicators. Growth in real terms, in volume of shares and in transactions, has been substantial but quite slower than invalue terms. At any rate, continental Europe is catching up with the Anglo-Saxon world, with-
out yet reaching the same degree of equity culture. This suggests that, apart from short term fluctuations, long term prospects of buoyant equities markets in Euroland are justified. Cornpetitiorz anzong stock exchanges is tlzerefore competition in exparzding markets, not in shrinki~zgones. e
Multiplicity and heterogeneity of stock exchanges
Each of the fifteen countries of the European Union has at least one organised stock exchange. Large countries often have regional exchanges, like the eight regional exchanges in Germany. National established exchanges compete witli dealer markets as NASDAQ and EASDAQ and try to integrate the derivatives markets in their own country. Additionally, they are setting up specific marltets for young and fast growing companies. Nouveaux Marchés (NM) are operating or planned in ten different countries, Germany being the outstanding example. Therefore, each analyst, studying the relations among the European equities markets, has to take int0 account the existence of three dozen European Stock Exchanges (Steil (1996), as quoted in Di Noia, op.cit., p.10). Exchanges in Europe are not only numerous; they are also quite different from each other. Each exchange has its own past, and its own historica1 roots which determine its generalprofile. In this respect, the case of the Benelux area is quite striking (Abraham (1999b)). This geographically limited area houses three exchanges which represent each a different type and tradition in European market places. Amsterdam has a tradition of self-organisation and selfregulation, following the Anglo-Saxon tradition. It maintains stroiig links with Wal1 Street and the City. Brussels has been strongly influenced by the French concept of a public institution with a Napoleonic and corporative organisation. The Luxemboiirg exchange, which is younger (1927), is rather a bourse de banquiers (Gillet et Minguet (1995)), following the German model. Three countries, three exchanges, three different models! However, these differences in genera1profile are now being reduced in the process of demutualisation: exchanges become enterprises and are less institutions, irrespective of their historical roots. Size is the most obvious factor of differentiation. In the various panels of Exhibit 4, London has mostly been chosen as the reference. In market capitalisation as wel1 as in turnover, the LSE is still about the
double of the Deutsche Börse. London tops Paris by about 140 % in market capitalisation and by about 40 % in turnover. Amsterdam has a market capitalisation of about one quarter, aiid Milano and Madrid one of about one fifth of London. But world dominance is still exerted by New York: the market capitalisation of the NYSE is still 4.7 times bigger than that of the LSE. In the Nineties, the genera1 trend has been towards reducing the differences in size. The growth of London has lagged behind the expansion of the two other main centres in the EU: Germany and Paris. The differente in growth is even more important vis-à-vis the exchanges in Southern Europe and most of al1 vis-à-vis Switzerland and Amsterdam. In general, London is considered to be on the defensive, while Frankfurt emerges as the rising star, after the localisation of the European Centra1 Bank in Frankfurt and the introduction of the euro. Differences in size determine differences in market power: This raises the question of leadership in Europe, either in competition or in cooperation, as wel1 as the fear of subordination of the smaller exchanges to the Big Brothers. Even more heterogeneity derives from differences in organisation, trading systems, procedures and regulations. (e.g. order-driven versus quote-driven trading sytems and procedures, screen-trading versus open-outcry, direct access or via middlemen, self-regulationversus external supervision, etc). In competition models, these differences can be used as an instrument of market power, the leading center trying to impose on or even to sell its trading system, its procedures and its regulations to the others. In models of cooperation, on the contrary, these differences are burdensome elements of non-compatibility, which can only be eliminated by strenuous efforts and negotiations. e
Interdependance
Everyday observation shows a definite convergence of the stock exchange indices at the European and even at the world level, under the dominant influence of Wal1 Street. However, the degree of convergence varies over time, reflecting specific developments and events in the various countries. The correlation matrices 1997-1999presented in Exhibit 5 show, for 1997, a high degree of correlation of most of the Euroland indices among each other, but also with the Dow Jones and the London FTSE. This is less the case in 1998 and in 1999, especially in the relations between Euroland on the one hand and New York and London on the
EXHIBIT 5 Correl~ztionïnutix,fo~ leading stock tnarket indices (1997, 1998, 1999 urzfil23.07) First periode: 01101197 - 31/12/97 Londuii
Gerrnariy
Pal-is
Amstzrdaiii
Urusscls
Pnii
Ain
Brussels
Biiis\cls
Milano
Mzidrid
I>iihon
Dow Iiriic\
Luxembourg
Milaiiu
Madiitl
Lishiiii
D~i\vJí~iics
Lusemh«iiip
Milano
Madiid
Lishi~n
Dtrv Jonc\
Luxein-
Gcim;iiiy P,iri\ Aniitcrdaiii Luucinhourg Milailil Madrid Dow Jiiiieh 0,93 Second peiiod: 01101198 - 31/12/98 Londoii
Li\hiiii 0 S34 Dow Jone\ 0.848 Third period: 01101199 - 23/07/99
Paiis Aiiisiei-daiii Biii$\cls I,uxcnilioui-2
0.828 (1,753 -0.617 0.755
11.713 0.810 -O,il5[! (1.685
(],SS7 -0.643 0.836
0.199 0.8f1n
-0 1198
other. It is certainly premature to attribute this lesser degree of interdependence to the introduction of the euro. Cyclical divergences between the two blocks provide a more convincing explanation. Howeve,: for thefuture, the question should be raised whether and how a successful cooperation in Euroland wo~ildchange the rather one-way interdependence with the American exchanges alzd whether this would reduce volatility in Euroland. s
Expectations
The introduction of the euro has given an electr(on)ic shock to the competition and the cooperation issues by the expectations it has aroused in the financial centres and exchanges. We have tried elsewhere to formalise these expectations in two diagrams, the first one describing the expectations about the capita1 markets in general, the other concentrating on the equities markets (Artesia (1999)). These diagrams are reproduced in Exhibits 6A and 6B. Most of the expected developments were already discussed in the preceding sections and exhibits. This presentation suggests another way of looking at the developments in the competition and cooperation among stock exchanges, i.e. by analysing whether and to what degree these expectations have materialised or are being changed in their confrontation with existing structures and interests. We wil1 use this approach, at least partially, in the subsequent sections.
EXHIBIT 6A
EXKIBIT 6B
Expectalions aboui rhe impaci of rhe euro ori capilal markers in gerreral
fipectaiions abour rhe specific inlpaci of the euro on srock niarkeis
J
I
lose segmentaiion and m o i e
opportunities of integration
l
insidc ihc stock exchanges world-
1
~ 1 ~ ~ I l - J higher atiraanienovr
concentration
reduciion of
EMU)
big financial centres
L__i Source: Arlesia
Bank
I
exchanges and more opportunity af direct financing . by markets
11
Bergers and acquisitionr and reduaion of Iisted shares -
1
B. Competition but little corzcentration The over-al1picture is quite striking; the expectation of increased competition has materialised but has not (yet) substantially reduced fragmentation of markets and exchanges. Large-scale concentration has not occurred. The easy reaction to that assessmeilt states that it is too early. A more genera1 explanation derives from the observation that in periods of expansion and prosperity on the securities markets, there is little incentive for exchanges to contract, to sell out or to merge. There is room for everybody. New low-cost participants enter the market because entry is easy and cheap. OTC (Over-The- Counter) transactions, in the broadest sense of the term, flourish and expand more rapidly than the business in organised markets. Dealer markets are very attractive for big individual transactions and for the listing of young, growing companies. The traditional exchanges, although on the defensive, can rely on the accuinulated stock of listings and relations. They react as suppliers of final liquidity and also by integrating companies, which have not only emerged, but matured and which can boast a satisfactory track record. In periods of tension and contraction the opposite developments often occur. How can such a scheme be applied to the recent financial globalisation in Europe? In recent years securities markets have been fueled by the prospect of the euro, by the need to finance huge industrial mergers and acquisitions, by the addition to the listing of newly privatised companies. New participants have entered the market, even through electronic trading. American intermediaries have discovered the potentialities of a unified European capita1 market. Established European stock exchanges have reacted by reducing prices and costs, by huge investments in electronic trading systems and by promoting remote access from abroad. As shown in Exhibit 7, they have been performing well, in this respect, in most continental countries, France being the outstanding exainple. While the investors'portfolios are being internationalised in the new perspective of Euroland, business in most national exchanges, except London, tends to concentrate ever more on domestic shares. In most national exchanges the share of listed foreign equities is declining. The opening of remote access, promoted by the European Directives, induces investors and intermediaries to operate on a security in the place where its liquidity is highest and the transaction cost lowest. This usually occurs in the country which is the home base of the company con-
EXHIBIT 7 Cost of executing trades (1996-1998, in ha sis point,^) Avci-agc commis~ions
UI< Buyiiig UI< Sclliiig Germany Fraiicc Switzcrland
15,9 21,4 21,6 33,O
5 22.2 22,O 23,s
Avcragc fccs
Ih,O 22,s 21,7 23,7
Sorrrce: Elkins arid MC Slici-ry (1998) and Financieel Ecoiioniische Tijd (1990)
cerned. As the international diversification of portfolios is strongly biased towards European blue chips, exchanges which provide the home base for such equities eiljoy a buoyant demand from both domestic and foreign investors. This is the case in most exchanges of the larger European countries but also e.g. in Amsterdam. For established organised exchanges, competition is hardest in the case of fast-growing high-tech companies of the mini-Microsoft type. As already mentioned, the Nouvea~lxMarchés have been set up as a reaction of the established exchanges against the challenge of the dealer exchanges and more specifically of NASDAQ and EASDAQ. At the moment of writing, five such marltets are already functioning (with varying success): Frankfurt (by far nôl), Paris, Amsterdam, Brussels, Milano. The admission of Stockholm, Switzerland, Copenhagen, Oslo and Helsinki is being considered. Another development, which raises more questions, is the pressure on established exchanges to lower their standards of admission and of trading. In its competition with the locally-based EASDAQ the Brussels Exchanges had to admit, immediately, to its PremierMarché, some companies, which had better be introduced in the Nouveau Marché. Amsterdam is presently considering to lower its standards of admission, which traditionally attached much importance to the track record of a company and hampered the admission of young, promising firms in this way. Summarising, present relations among European exchanges feature new participants, strong competitive pressure, actual competition from dealer markets, virtual competition from ECNs -, priceand cost-cutting and some relaxation of standards in the established stock exchanges. On the whole, structural changes remain quite limited. A first one concerns the merger of the various exchanges and ancillary services in the samefinancial centre. These mergers aim at setting up a one-stop organisation in that financial centre, and to foster the competitiveness of the over-al1 exchange and of the financial centre itself. Typical examples are Deutsche Börse (1993), Amsterdam Exchanges AEX (1997), Brussels Exchanges (1999), Paris-Bourse-SBF (1999). In these cases technica1 progress makes for bundling of functions and services under the pressure of competition. Demutualisation is the second significant structural change. As mentioned above, it transforms a corporative institution, working on a nonprofit basis for its members, into a firm, an enterprise, with its own
capita1 basis, its own bottom line, and sometimes also its own listing on the stock exchange. Lefebvre (1999) rightly stresses the fact that demutualisation first started in the exchanges (Stockholm, Helsinki, Sydney), which were most under coinpetitive pressure. The strongest resistance against demutualisation came, until very recently, from the big dominant markets: London and New Uork. But als0 there, things are changing very rapidly: now, the principle of demutualisation seems recognised in both centres. Concluding, we may assert that, under present competitive conditions, the huge problems to be faced and solved by European exchanges remain relatively unchanged. They have been hidden by the boom in the equities markets in the late Nineties, but will reappear and become more pressing when a recession temporarily interrupts the long term expansion of the equities markets in Europe. Then, the fundamentals of the European situation will come to the fore: B
e
e
in the continental exchanges, technological advance but fraginentation of market places; in OTC dealings and in dealer markets: dependence on mainly Anglo-Saxon based systems; over-exchange-trading (analogy with over-banking): too many exchanges chasing too few stocks which are sufficiently representative, well-known and well-analysed at the European level; difficulties in adapting the institutional framework and the prudential control, which remains largely decentralised, to the needs of inarkets, which are becoming increasingly globalised; potential delocalisation of capitalisation and turnover because of weakening links between transaction, exchange and financial centre (Artesia (1999)).
C. Cooperation: Agreements, Alliances, Misalliances It is common knowledge in economics that whenever competition gets tougher and threatens profitability, calls for agreements and alliances become more frequent and more insisting. This also applies to stock exchanges. These agreements have obvious implications of market power, by accumulating or dividing market power. In this respect, they should not leave competition policy indifferent.
However, this does not mean that the present flourishing of agreements and cooperation proposals can be reduced to mere attempts of cartellisation or to soine form of imperfect competition. There are other other economic aspects which should also be taken into account, especially in Euroland. In the context of financial globalisation, economies through the use of large, technologically advanced electronic trading systems are of utmost importance. Agreements and alliances may be indispensable to enjoy full economies of scale and to transform technological progress int0 competitive edge (cfr the SAX-2000 example in Part 11). On the other hand, network economics highlights the externalities resultilig from setting up and extending networks2. Moreover, if mergers and acquisitions are not feasible or even advisable, agreements and alliances can, at least partially, substitute for them (as implicit mergers) and enlarge existing but fragmented networks. Finally, from the economic policy viewpoint and from that of prudential control, the public authorities very often prefer agreements and alliances to full (cross-border) mergers because they usually involve less delocalisation and dilution of value added, employment and prudential control. Some agreements are even considered necessary as a matter of public interest. When externalities exist, competition may tend to monopoly, which process may be better controlled in the case of agreements and alliances. In the present state of affairs, three broad categories of agreements can be distinguished. Specific agreements on a well-defined and limited topic are the least ambitious and the most frequently practised device. Such agreements are particularly numerous on the derivatives markets (for an extensive list, see Ansidei (1999)). The most recent ones, which are planned to become competing transatlantic alliances, are the agreements Chicago Board of Trade (CB0T)-Eurex and Chicago Mercantile Exchange (CME)-LIFFE (London International Financial Futures and Options Market). Other agreements involve sales of trading systems, which may, aftenvards, lead to networks with more or less compatible trading systems. In Europe, two systems predominate in this respect: the N(ouveau) S(ystème) de C(otation)-NSC, promoted by SBF (France) and sold world-wide to about twenty exchanges and XETRA, promoted by Deutsche Börse and sold to 5 other exchanges, mostly in German speaking countries (Ansidei (1999)). These agreements are, at least in the beginning, limited in scope and in time horizon. Being fragile, they often disappear before being
implemented, or they are replaced by another agreement, often with another partner. If they are successful, they usually result in an alliance of a broader scope and of a more permanent nature. Agreements leading to concentration through fgderation, acquisitions or rnergers are, of course the stroligest form of cooperation. They usually occur under the leadership of a dominant partner as shown by EUREX (1996), which merged the German and the Swiss derivatives markets under the leadership of Deutsche Börse. In the field of cash stock exchanges, examples of such concentrations are, up to now, limited to mergers between regional exchanges (e.g. the announced merger of the Hamburg and the Hanover exchanges int0 the North German Bourse) and among various types of exchanges and services in the same financial centre (cfr supra, section II1.B). Alliances integrate specific agreements in a broader and more permanent framework. They are, at present, the royal roads by which stock exchanges try to overcome the existing fragmentation and, in Euroland, to meet the challenge of an integrated financial market with a single currency. Two main alternative models have been proposed as basic schemes and guidelines for setting up such an integrated system. The model of interconnectivityamong existing exchanges is strongly advocated by the smaller-sized exchanges in the Benelux area and in Southern Europe. At the other end, the concept of a single pan-European platform has been jointly launched by London and Frankfurt, as the centralised trading place of 300 European blue chips (July 1998). The main characteristics of each model, with its strengths and weaknesses, are presented in Exhibit 8, which is based on the Artesia Bank study. Interconnectivity favours a progressive approach based on subsidiarity and pays due respect to the specific identity of each participating exchange. In this model, the exchange is still near the local investor. But interconnection allows him through remote access and crossmembership to have his transaction made on the exchange which offers the highest liquidity and the lowest cost. However, the investor has to conform to the organisation and the procedures and rules of that exchange, which means time, energy and costs. To overcome this obstacle, the heterogeneity of organisation and regulations has to be reduced among the participating exchanges. Experience shows how slowly and with how much difficulty harmonisation proceeds in Europe.
EXIIIBIT S Alfernat~venlodels of z~ztegratedsytienl of E ~ ~ i - o l ~ ex(-l~arzger ear~
1
INTERCONNECTTVI1 Y MODFI
PAN kUI~.OIJLANP L A T t O R M MODtL
UASIC PRINCIPLE
'l'lic cxistiiig sicick cxcliariges nrc coiineclerl witli eacb «[her. Eacli ineinlxi- lias access io tlic »latforrri u i ihc otlicr
Coinmoii platform rif (100) top-Eui-~~]ie;iii shaics
I>F,C;AL SïflT'US
Cioss meriihcrsliip
Coiriiiion isoiiipariy owiied liy pat-iicili:itiiig excli;irigcs
LOCf\I.ISATION
Iii Siiiaiici;il ceiilei- nf eiieli pal-licipating sluck exchniigc
Iii doniiiiaiit liiiancial ceiitei «r oii Lioiiic IIC~>N«I-I< -~
STRENGTHS
-
-
-
WEAIWESSLIS
-
-
-
;i
1
\~islu;il1i;isis as a n elcc-
~
'li-aiiiacii»iis can Re iiiatic at Llie stock cxch:iiigc witli tlie Iiighcst liquitlity aiid tliz Iowest cost R-ansactioiis ai-e made witli diic I-cspecl lor divci-sity of orgtiiiisation arid I-cgiil:itioiis in llic cxisliiig stocli exchaiigcs I.inki willi lina~ici;ilcenti-es oS origiri ;irid local clie~ilele;irc rii;iirilaiiicti: rio del~icalis;~ti»nof valne ;~cltlcclantl enipluyinclit tlarinoiiisatiori ~ i 01-gaiiisation t and regulntions c;iri Iie realiscd grndu;illy Eacli exchaiigc Lccps a divei-sificcl Iisting «f blue ellips. miclcaps aiid siii:ill caps Iii c(1iiipctition witli Big Hiotlici-s. owii ideiiiiiy «i siii;iller exchanges is sccurcd
l a r g e li~luiditypool is \ct up for Ilie I-ieil Eiircipcnii sliai-ci Optimiialioii of tccliiiic;il ~~pporliiiiitiei niiil c c ~ ~ i i t ~ r i111. ~ics sc;ilc witli concc~iiiitaiilreductii~iiof pi-nduclii~iiaiid !r;isii;iciiciri cosls - Homogcneily » i »rgariis;ilioii aiid icgulati«ii\ 1.01- Europcari tiliic chips - Optiinisetiun iiSiiitroducti«ii «ïciiio as :I c;italys«i- ol'clinngc - 1iriporl;iiii word-widc effecls «n Siiiaricial ni;ri-licis
No Sul1 pociiiiig «i liquidity ?rnni:icti~iii Iias to bc inade aiitl scttlcd aceui-diiig icgulatiritis aiid pi-ocedures of cacli excliaiige witli col-responding COSIS Leas iiiteriiati«nal cSlccts ELIIO-cftcctis siiore Iiiiiitcd: pulicy CISm a i l stcpi witlir~iil c«mlirclicilsive plaiiiiing
-
-
Abrupl ceii1i:ilisatioii iieglectiiig thc suh\icii;irily priiiciplc If I ~ i ~ i l i i ~inc d«min;iiit d fiiiniicial center. «I-gnriisatii~n aiid I-cgulaii«iis wil1 probahly been hascd «i1 local cuiiciiti~ins - Iri the uIlici-cxchaiiges. lisliiigs wil1 he lirnited 10 1oc;il slini-c. prcd«niinaiilly r11 incdiuiii 01- iin:ill-\i~cd c«inp;iiiic\ - Earriiiips il-iirii I~iisiricsswil1 tie s-rpl:iccd iiy dividcridi; los5 o1 local value ;idded aiid iohs - Ar1iSici;il firiarici;il sliiiiiilus 1.01- c~,mpaiiicst
The pan-European model is radical in its scope, with al1 the consequences, good and bad ones, of radicalism: full use of the opportunities provided by technological progress and European integration, creation of a large pool of European exchange liquidity, impressive worldwide effects. Less appealiilg is the displacement of trading aild employment, resulting froin the association of the platform with a domiilant centre. When, in July 1998, the alliance Londen-Frankfurt was announced, reactions were rather emotional on the side of the main centres, which, as Paris, had not been approached, and of the smaller-sized exchanges, which were already discussing cross- memberships according to the interconnectivity model. The idea of a counter-attack by setting up a rival group was rapidly abandoned and replaced by a strategy of ifyou can't beat them, join thern! A club of eight exchanges emerged, grouping London, Frankfurt, Paris, Zurich, Amsterdam, Brussels, Milano and Madrid. On 4'" May 1999, a statement of intention, signed by the eight members but open to others (Lisbon has already applied) was issued, announcing the constitution of a pan-European exchange, based on the model of an electronic platform of European blue chips, with unified regulations. Many technical, regulatory and fiscal problems are to be solved, which explains that the start cannot be expected before the end of the year 2000. Has the point of no return been reached by that statement? Specialist~of European integration will probably be rather cautious, remembering the ill-fated project of the unified statutes of the European company, which, later, had to be replaced by the mutual recognition of the national legislations. Anyway, the measures decided up to now to implement the plan are quite modest. Trading hours have been unified between London and Frankfurt (9 a.m. till 5.30 p.in. ). This measure has been followed, at least partially, by other members of the club (e.g. Brussels: 9.30 a.m. till 5.00 p.m). A more significant step is the reciprocal opening up of the access in Frankfurt and in London, in order to have transactions on German stocks made directly from London in Frankfurt and vice versa. Oddly enough, these measures proceed more from the interconnectivity than from the pan-European model. The basic problems for implernenting the latter are far from solved. They concern the choice of the trading system-XETRA or CREST- and most of al1 the ownership of the entity (who will be the most influential shareholder, the
leader of the common company?). Ruinor about a possible failure of the project has spread in the press but this has not been confirmed. At any rate, olie conclusion emerges: thepoint ofno return has rzotyet been reached"8 Meanwhile, the initiatives integrating interconnectivity principles have multiplied, in spite of the negotiations on the pan-European platform. They reflect geographical, cultural and financial proximity and concentrate on remote access and cross-membership. The two most significant projects are: 0
The Benelux alliance, which started in January 1999, after the internal functional mergers AEX and BXS (cfr. the institutional and legal analysis of the scheme in Taevernier and Pycke (1999)). Starting from cross-membership and cross-access, it now progresses towards a network without cross-border interruption in trading, clearing and regulation; The alliance Paris-Ziirich-Milano, complemented by a Paris-Lisbon agreement. This alliance is also based on cross- membership and cross- access.
The chronological parallelism between the pan-European platform negotiations and several interconnectivity initiatives is undoubtedly confusing. It suggests that several exchanges develop interconnectivity relations which may substitute for the platform, if the latter fails to materialise. In that case linlting tlie Benelux and the Paris-ZiirichMilan schemes and adding other smaller European exchanges (e.g. Lisbon and Athens) will result in a European system which favours interconnectivity and rejects centralisation. But participation in the two models is not an adventure without risks, especially for the smaller-sized exchanges. If the single pan-European platform prevails, the profitability of many interconnectivity relations may be substantially impaired, because the most profitable part of the trading business will flow to the single platform at the expense of the interconnections. In Scandinavia at least, two exchanges-Stockholm and Copenhagenhave made up their mind and Oslo has recently decided to join them. These exchanges have decided to stay out of the club of eight and to set up their own scheme: NOREX. This alliance is based on crossmembership but also provides for sharing the same trading sytem (the famous SAX-2000!) and the same trading rules, concentrating liquidiîy in one order-book per security (Skaanning-Jörgensen (1999)). The
participant exchanges are shareholders in a jointly owned company, which transforms Norex iilto an implicit merger or, at least, in a single mini-Scandinaviarz platform.
IV. CONCLUDING REMARKS: STRUCTURAL CHANGE THROUGH THE MARKET O R THROUGH ORGANISED REFORM? A. A confusing over-al1 picture
Competition and cooperation among European exchanges is a fascinating issue in the context of financial globalisatioil. It is the outcome of the dialectic interplay of technological progress, de- and reregulation, European integration and the penetration of the equity culture and other socio-economic forces. However, our empirica1 survey shows that, at the present stage, the over-al1 picture is quite confusing, even chaotic: it is a mixture of heavy competition under the pressure of financial globalisation, and, sometimes Byzantine, discussions about alternative and alternating schemes of cooperation, without much effective structural change. The original idea was ambitious and straightfonvard. In the context of the introduction of the euro, it aimed to set up an integrated system of European exchanges, to meet the challenge of an enlarged European equities market without exchange risks and costs. In a certain sense, this integrated system of European exchanges would parallel the integrated system of centra1 banks, which is now operating under Stage Three of the European Monetary Union. Our survey shows that this basic idea was easy to formulate but much more difficult to iinpleinent by private competitors, without much political support. As time goes on, the original impetus coming from the introduction of the euro may weaken and leave the present situation of extreme fragmentation largely unchanged. In this hypothesis, the final solution will be given by the markzet. Unless organised official exchanges integrate the ilew developments in electronic trading and networlts very rapidly, it is not very probable that they will be on the winning side, in a context where technica1 progress and networking induce a cornpetition with externalities and the ultimate survival of the fittest.
B. Pleadirzg for a well-balanced organised reform ''In this electronic age, the physical location of the trading forum is of relatively minor irnportarzce:an electronic exchange could be located on an ice-brealceron the North Pole. That does not make the Arctic tlze centre of the share trading business. What matters in terms of geography is the physical location of the people providing tlze fïnarzcial services: the brokers, dealers, professional investors and analysts working around the secondary nzarket, as wel1 as the bankers, underwriters etc, who service the primary market. And it is not just the location of the jïnancial intermediaries that is at stake. A larger issue is the performance of thefinancial services industly as a contributor to the real economy. How eficiently are these services beingprovided? How innovatively? How wel1 regulated are they? How is al1 this affected by intensified cross-border competition?" (Guido Ferrarini (1998)). In our opinion, the final outcome of the present interplay between competition and cooperation among European exchanges should not be left to the mere interaction of market forces. Opportunities to proinote and channel structural change by a well-balanced organised reform still subsist. Exchanges, even after becoming enterprises, perform some public functions, bringing together supply and demand of capita1 in ai1 orderly way, and contributing significantly to value added and employment in a city (e.g. London), a region (increasingly, Frankfurt and Brussels) or even a country (e.g. Luxembourg). As emphasised in the quotation just given, the basic issue is not about technical solutions but about the future of the financial communities and the financial centres in different countries, large and smal1ones, which, together, are building the European Union and more specifically the Euro-zone. In order to reach a large consensus on an organised reform, extreine solutioils should be discarded. The present, rather extremefragmentation is not sustainable in the long term. Too many exchanges are chasing too few representative European stocks. Maiiitaining the present situation, means slow degeneration for the smaller-sized exchanges and recluced opportunities of expansion in the larger ones. Extreme centralisation ignores the link with the local financial coinmunities in the several meinber countries, the link with the local investor, the local firms on their search of risk capital, the local intermediaries - the bankers, the brokers, the analysts - the national and local public authorities, in charge of promotion and supervision. In
general, extreme centralisation is based on the wrong hypothesis that the single European equities market will be a uniform market, just at a moment when more and more small and medium-sized firms, without much international recogiiition, want to have their growth to be financed through the exchanges ratlier than by the banks. If both extreme fragmentation and extreme centralisation must be avoided, the next question is: along which lines should an intermediate scheine be designed as a blueprint for an organised reform, based on a large consensus, both from the niarket participants and from the public authorities? The choice between interconnectivity and single platform alliances may be, in the long run, of minor importance as the basis of the reform. The experience oí'many financial alliances suggests that their outcome is quite uncertain: either they dissolve or are dissolved and disappear, or they spontaneously direct the most significant transactions towards the most performant centre, in which case there is no longer a difference between interconnectivity and single platform. In our opinion, an adequate reform should aim at setting up an integrated system promoting a division of labour distinguishing the market for international blue chips, the inarkets for shares with local recognition, especially small and midcaps and the markets for fast growers. This distinction already exists at the national level and should now be applied and extended at the European level, with its implications on the (re)structuring of existing exchanges. We may conceive the future integrated system of European exchanges, which should emerge from the present interplay between coinpetition and cooperation, not as a perpetuation of the present dispersion, nor als a inonolitliic institution in a dominant centre, but as a system of three concentric circles, distinctive but interdependent. In the inner circle, at the centre of centres, a single pan-European platform, composed of a server, software, procedures, regulations, information and public relations. The physical components of the platform will be located in a dominant centre and be strongly influenced by local conditions in that centre (as the ECB is influenced by local conditions in Frankfurt). On the contrary, most contacts with the investors, the listed companies, and many intermediaries will remain decentralised in the various participating countries and exchanges. A second circle should mainly comprise the Premiers Marchés of the present national exchanges. When the international blue chips in these Premiers Marchés have left for the inner European circle, the exchang-
es of this second circle will secure liquidity plus quality for shares with regional or national recognition, including, be it not exclusively, small and midcaps. In the smaller countries, the subsisting national market will not be large enough to justify an exclusively national exchange. Middle-sized exchanges in neighbouring countries should engage in promoting supra-regional integration of their regional markets, particularly of their emerging markcts of small and midcaps. Followiilg the example of NOREX, they should also strenghten the institutional framework of their regional cooperation by setting up common institutions, the final outcome possibly being some sort of merger. In a liistorical perspective, this development would echo the gradual decline of regional exchanges and the growth of a national exchange in most European countries. The third circle, as it often happens in a system of concentric circles, looks more foggy and confusing. lit concerns the market for fast and innovating growers, where Nouveaux Marchés(NM) and dealer markets, as NASDAQ and EASDAQ, are competing to list and to attract the business of the Same companies. Experts (Jean Peterbroeck (1999)) strongly advocate the merger of NMs and EASDAQ, arguing that also markets of growing companies need liquidity and stability. This aim should best be reached by merging NMs and EASDAQ. We are not yet convinced by this approach. The history of the Nineties has shown how inuch the financial communities in continental Europe have benefited from the competition between the existing auction markets and SEAQ-I, the prototype of a dealer market. On the established exchanges, commissions have been liberalised, costs reduced, orders centralised, automation extended. An analogous effect could materialise from the competition between shares, cross-border listed on EASDAQ and NASDAQ on the one hand, and on a more national NM on the other. In general, we believe that an integrated system of exchanges would ultimately benefit from the sting of coinpetition coming, at the margin, from dealer markets and ECNs. It would also stimulate the NMs in the various countries to intensify their integration, in order to build a real network of cross-border services for fast growers. The scheme of the three concentric circles, which we have developed in our concluding remarks, does not result from an abstract intellectual exercise in the ivory tower of a studies department or a university centre. It is an effort, at the present somewhat confusing stage of the competition and the cooperation among European exchanges,
to combine the most promising but sometimes dispersed proposals int0 a more systematic setting, which would, in our opinion, quite adequately meet the challenge of a unified, but not uniform European capita1 market with a single currency. NOTES 1. Thc present essay has been prepai-cd for the 1999 Annual Meeting oí'the European Associatioil of University Teachers in Banking aiid Financc (Lisbon, September 1999). It draws on a presentation made at tlie Financial Forum Antwerp o11 tlle topic: Firz~~ncial Glohnlisution. Its impact on ihe E ~ ~ r o p e aStoclc n Exchanges. with syecinl derence to tlze Belzelirx-Aren (April 1999). Tlic specilïc problems of the Benelux cxcl-iariges arc discussed in a separate paper, to be publishcd in Revuc de la Banque/Banl<- en Financiewezen. Whcn preparing tlie present text, the authors integi-ated parts oí'their study for the June 1999 issue (n" 200) of thc Notes EconomiquesIEconoinisclie Berichten of Artcsia Bank on the impact of tlic euro on the capital marlcets in Europe. 2. Di Noia (1998) discusses two games wit11 iietwork exterria1itics:The first model featurcs competition among incomnpatible exchanges, thc iilcompatibility arising froin differcnces in telinology, tradilig systeins, regulations etc. A firm expected to have a large network, prefcrs incompatibility and cornpetition between incompatible exchanges. Exchanges that are dominant and have decisive cost advantages may avoid agreements aiid implicit mergers as they wil1 be, i11any case, the only to survive. The second model is o11e of complete compalibility, featuring implicit merger (eacl-i exchange lists tlie fii-ms,which are already listed in the partner exchange) and remote access (granted wit11 reciprocily and without additional cost to the members of the partner exchange). This model does not induce Darwiilistic competition.1mplicit merger always improves welfare,when there are cross-advantages i11marginal costs of the participating exchanges. But in the absence of any coordination or policy guide,implicit merger does not arise spontaiieously. Implicit mergel- is a clearstïategic option lor exchanges. These two extreme cases highlight the difference behveen solutions by merc nlarlcel foi-ces and those reached by coordination and cooperation.
Abraham J.-P., 1999a, Le strategie delle banche iii Eurolaiidia iil Banca, Iinpresa, Societa', April 1999, 3-34. English version as Uiscussion Paper DPS 99105, Center for Ecoiiomic Studies, I
Di Noia C., 1998, Coinpetition and Integration amoiig Stock Excliringes i11 Europe:Network Effccts. Implicit Mcrgcrs and Rernote Access, Wliartoii Financial Institutions Center Working Paper Series 98-03, 48 pp. Ecoiloniides N., 1993, Nehvorlt economics with Application to Finaricc i11 Financial Mai-kcts, Irlstili~tions& bistri~rnci~ts 2, Dccciiiber, 89-97. Ferrariiii, G., ed., 1998, Europeari Securities Markets, The Iiivesliiient Serviccs Directive aiid Beyoricl, (Kluwer Law International, Londoil-The Hague-Boston), 213. Gillet R., Minguet A., 1995, Mici-o-structure el rénovation des inarcliés financiers cri ELIrope, (Paris: Presscs Univcrsitaires de France), 420 pp. Iiistit~itoper la Ricerca Sociale (IRS), 1999, La Riorganizzaiiionc dei Mercati in Rapporto IRS sul Mcrcato Azionario 1999, 143-177. Interiialional Monctary Fund (IMF), 1997, Glol>alization,Opportunilies and Cliallciiges, World Ecorzonzic Oi~ilook,May, Chaptcr 111. Intcrnatio~ialMoiietary Fuiid (IMF), 1998, Globalization of Finance and Fiilancial Risks in Interliational Capital Marltets: Developmeiits, Pi-ospects aiid ISey Policy Issucs, AIIiiex V. Jacquillat B., Gresse C., Gillet li., 1998, Revisitiiig the Cornpctition betweeii tlie International Equity Marltet of Lhe London Stock Exchange aiid the Br~isselsStock Exchange, Revite de li( Bniiq~~eIRn~lketl Firiiiiicieweze~~, September, 375-383. 1Sane E.J., 1981, Acceleratiilg Inflation, tcclinological innovalion and the dccreasiiig eí'í'ectiveness of bariking regulation, Jo~ifliuIofFilzance, May, 355-367. Lefebvre O., 1999, Bourses: cntre fusions el alliailccs, Dossier: Bezirsalli~irztie.r/Aliiariceshoi~rsieres, op.cit., 251-254. Official docuineiits (annual reports, rnonthly bulletins, press releases and statistics) of 111ternatioiial Federation of Stock Exchaiiges (FIBV), Amsterdam Excliaiiges, Brilssels Exchanges and Bourse de Luxembourg. Onado M., 1999, The conseqilences of European Fiilancial Integration for the Regulatory Authorities, Leuven, Geiieral Bank Cliair Lecture, March 1999 (forthcoming). Pagano M., 1998, The Changiiig Microstructure of European Equity Markets in Ferrarini G,, ed., European Securities Markets, Tlie Investineiit Services Directive and Beyond, (Kiuwer Law Interilatioilal), 177-204. Peterbrocck J., 1999, Bruxelles nc doit pas craiildre la création d'uiie Bourse européennc, Brussels lecture of 27.05.99, as reported by L'Echo of 28.05.99. Press information and comments in De Financieel-Economische Tijd, Het Financieele Dagblad, L'Ecbo, The Financial Times, Tlie Wall Street Journal. Röell A., 1998, Competition among Eilropean Exchanges: Recent Developments in Ferrarini, G,, cd., Europeaii Securities Markcts, Llie Investment Services Directive and Beyond, (ISluwer Law International), 213-2211, Skaanniiig-Jorgenson P.E., 1999, Norex-Nordic Excliangcs, Dossier: Be~~rsnlliarzties/Allin~zces boilrsieres, op.cit., 266-267. Tdevernicr B. aiid Pycke A-S., 1999, Cross Access:een raak antwoord voor de toeltomst van Briissels Exchanges, Dossier Be~ilsallicrntieslnllin~ices boi~lsieres,op.cit., 255-265. Walter I., 1998, Globalizatioii ofMarl<etsand Fiiiancial Center Compctition, miineo, March, 35. Wall Streel Joiirrial, 15.03.1999. Williarnson C., 1999, Structural Clianges in Exchange-tradcd Markcts in Bank oJ'Englarld Qi~alerlyBulletin, May, 202-206.
APPENDIX Key fgures for stock exchanges
EXHIBIT 4A Key figures for rnain European f rznncial centres
Concentration of domcstic turnover (1998) (%l (3) Index volatilily (daily clianges in Avcrage witli standard error hctwcen brackets 1997 1998 1999 (unlil 23107199)
59.8
85.5
63.4
0.72 (0.95) 0.98 (1.31) 188 (1.4)
1.08 (1.44) 1.39 (1.91) 1.08 (1.46)
1.01 (1.37) 1.18 (1.62) 0.96 (1.27)
72.1
%j)
(l) TSV: Trading Systein Vicw: Transactioiis on the Stock Exchange. (2) REV: Regulatory Environment View: Transactions oii and off tlie exchange under tlie supervision of tlie marltet authorities. (3) Share of thc 5% most capitalised listed companies.
Source: FIBV and Artcsia Bank.
EXHIBIT 4B Key fig~~res for Benelux excha~zges
R;itio ninrket vaiiicIGDP
domestic compaiiies (1998) billion EUR Value of share trading (1998) hillion EUR TSV registration REV rcgistratiori 1990=100 TSV registration REV registration Turnuver velocity of domestic shares (1998) Concentration of domestic markel value (1998) Concciitretion of domcstic turi~ovcr (1998) (%J) Indexvolatility (daily changes in %) Average wiih slandard error belwcen brackets 1997 1998 1999 (until23107199)
34
2
3
346
52
2
921
597
677
7L.4
27.8
4.2
73.3
56.5
33.1
67.1
54.9
61.9
1.10 (1.48) 1.32 (1.75) 0.99 (1.33)
0.77 (1.04) 0.94 (1.23) 0.79 (1.03)
0.44 (0.74) 0.51 (0.80) 0.59 (0.90)
EXHIBIT 4C Key figuïes foï Southein Euïopean exchanges Stock markelf irnportancc in lhe nalioiinl cconomy R;ilio m:iikei valuc/CDP 1997 (%) Markct capita1ia;ilioii«Isliaresof domeaiic compaiiies (April 1999) biilioii EUR 1990=100 Londen= 100 Numhcr of compaiiies witli sliarcs listed (April 1999) Total of which domeaiic Numbcr o i companies ncwly lislcd (1998) Toial of whicli doniestic coiiipanies Gross aiiiount of new capilal rniscd by domestic conipanies (1998) biliion EUR Valoe oishare Iradiiig (1998) billion EUR TSV registralion REV regisIraiion 1990=100 TSV regiatraiion REV registration Turnover vclocity of domestic sliarcs (1998) Conceiitration of domestic markelvalue (1998) Concciitration «S domcstic iurnover (1998) (%j Iiidex volaiility (daily changes in %) Avcragc with slandard error beiweeii hiackcts i997 1498 1999 (unlil23107199)
I'rALY
MADRID
LISBON
30.1
54.7
36.4
521 369.8 22.9
364 346.8 16.0
50
243 239
527 523
131 131
22 21
113 111
S 8
10
l (l
3
417
547
41
102.3
170.3
82.2
59.6
66.9
52.8
60.0
93.5
59.4
1.09 (1.45) 1.58 (2.10) l .O3 (1.43)
1.02 (1.34) l .39 (1.92)
0.66 (1.01) 1.12 (i.59) 0.75 (1.07)
Z
1088
0.97 (i ,411
EXHIBIT 4D Key Jgures for referente US exchanges NYSE
NASDAQ
132.7
137.7
Stock markcts imporiance iii tlic iiaiional ccoiioiny Ratio niarket valuc/GDP 1997 (4,) Markct capiialisaii«n olsharcs of domcsiic compaiiies (April 1999) hiiliun E U R 1990=100 L«ndoii= 100
10,073 419.5 469.7
Niiiiihcr o l companics witli sliarea lisicd (April 1999) Tuial of wliicli domestic
2647 22h2
Number of companics ncwly liaied (1998) Toial of which domcstic coiiipaniea
205 162
Gross aiiioiiiii of new capital raised by clomcsiic companies (1998) billion EUR Value of share i r a d i n ~(19%) billion E U R TSV regisirdtiun R E V registratioii 1990=100 TSV registratioii REV registration Turnover velociv of dornestic shares (1998) Concentralioii of doniestic markci valuc (1998) Concentratioii of domeslic tiiriiovci(1 998) (l%) Iiidcx volatilily (claily cliatiges iii %) Average willi standai-d errorbciwccii hrackcis 1997 I998 1999 (iiiiiil 23107199)
487 437
29
l34 ~
6255
-
~p
4717
552 69.9
257.7
63.8
75.2
51.4
78.8
0.84 (1.10) 0.88 (1.23) 0.79 (1.02)
Tijdschrift voor Economie e n Management Vol. XLIV, 3, 1999
Principles of Corporate Governance with an Applieation to the Financial Sector by C. VAN HULLE*
I. INTRODUCTION For the last several years the issue of corporate governance (= the problem of the design of performing decision structures at the top of firms) has been heavily debated in Belgium and in most other European countries. Everywliere, these discussions focus on the question whether or not the countiy's current governance model should be left or adjusted in favour of the US (Anglo-Saxon) approach to the subject. Ever since the acquisition of the Generale Maatschappij by the French holding group Compagnie de Suez, the take over of large domestic firms by foreign companies has been a bone of contention in Belgium. Also the fact that few Belgian companies have been able to grow int0 large multinational firms has fuelled the discussion about the Belgian governance system. The current governance debate concerns the relatianships behveen share holders, board of directors and management in publicly quoted firms. At least at first sight, the structures that govern these relationships are remarkably similar between US and Belgian firms. In both countries the duties of the general meeting of share holders consist of yearly returning issues like approval of the annual accounts,as wel1 as exceptional decisions like approval of adjustments in the corporate charter. Next to the possibility to vote, share holders also have the right to receive dividends. In both countries the general meeting appoints the board of directors that in turn hires and fires management. X;
Departement Toegepaste Economische wetenschappen, K.U.Leuven, Leuven.
Notwithstanding preceding similarities, these structures actually seem to f~inctionin a vely different way. Particularly, in the US share ownership in publicly quoted firms teilds to be relatively dispersed and management holds a very powerful position vis-à-vis the board of directors. In particular, the board consists of both inside directors (= representatives of management) and outsiders (= have no link with the firm nor its management and usually do not represent large share holders) and its role is mainly confined to monitoring management. Mowever in practice management often exercises considerable influence on the board. In terins of balaiices of power, the Belgian situation is quite different. Ownersliip is concentrated and, next to reserving some seats for the management team, usually the dominant owners distribute board seats ainong themselves according to ownership proportions. Directors that are independent of management and large owners play a minor role. As these latter equity holders doininate the board, they actually hire and fire the management and strongly influence the firm's strategic choices. Although their influence has been shrinking over the last decade, a class of large owners that still plays an important role in corporate control are holding groups. However as large holding groups typically exercise control over several (many) firms, in a publicly quoted subsidiary conflicts of interest may easily arise between the parent and the smal1 share holders. For the former wishes to further the interests of the group as a whole while the latter prefer that the company in which they have invested thrives. This is very different from the US where holding firms virtually are non existent. The main theme of this paper concerns a discussion of the AngloSaxon governance system in comparison wit11 Belgian governance from the perspective of transaction cost economics. In the wake of the pioneering work of Williainson ((1984), (1985)) this approach has been very successful in illuminating many difficult governance problems and in pinpointing the role of important governance devices like the board of directors. It also allows for a clear analysis of the famous and competing share holder and stake holder models. This is very useful, especially in Europe where countries are still struggling with moulding or remoulding corporate governance structures, and where the advantages and disadvantages of both latter models continue to be a matter of debate. Furthermore it wil1 be claimed that in the wake of the continuing integration of European stock markets (some) convergence of the Belgian system to more Anglo-Saxon governance habits
will be unavoidable. Given this adjustment, the question then arises to what extend these changes in governance may affect financial markets and hence the business of the (Belgian) financial firms. A second and related theine of the paper is the interaction between regulation and corporate governance in the financial sector. In particular, both regulation and governance are concerned with resolving problems of conflicts of interest. Hence, the question whether the analysis of governance problems could lead to useful insights for regulation. It will be seen that when applied to specific regulatory issues it indeed adds useful perspectives. Section I1 of the paper describes the development of the share holder and stake holder models. Section 111respectively IV discuss the logic of the US (Anglo-Saxon) and Belgian (European) governance models. Next, section V considers the reasons why the Anglo-Saxon model is so influential in the European governance discussion. Section V1 contains some remarks on the question whether or not corporate governance will transform the Belgian (European) financial sector. Section VII considers two specific regulatory issues and discusses additional perspectives from the governance analysis. Finally section VIII offers some concluding remarks. II. GOVERNANCE MODEL§: THEORY Theoretica1 models of corporate goverilance are concerned with the design of efficient decision structures at the top of firms. These decision structures comprise the relationships between share holders, managers and other important parties that are involved with the firm (employees, suppliers, customers), ... As the issue at hand essentially is an organisational problem, the first approach would be to set up wel1 balanced organisational structures. Such structures should possess properties like separation of roles to create checks and balances, a clear set of rules of conduct, checks that these rules are obeyed and overall transparency. However applying these ideas is not so obvious in the case of corporate governance. In particular from an organisational perspective there exists an important difference between the top of the firm and al1 hierarchical layers below. Specifically, al1 members in lower layers have a 'boss' who serves hislher own interests by keeping a check on hislher subordinates. This is not the case for the top level, and, as will be seenbelow, organising checks and balances with respect to this layer is very difficult. In fact this latter problem is
the core issue of the governance discussions al1 over the world. As it wil turn out, no ready made solutions exist and that any choice has both advantages and disadvantages. Quite a iiumber of issues have to be tackled. For example, if there are many different share holders, as is typically the case in publicly quoted firms, important questions arise about how the relationship between the owners and the management sliould be organised. Should it be through a special body and if so what should be its tasks? And how should the relationship between employees management and share holders be set up in the structure? Two different models, the share holder and the stake holder model provide an answer to a number of important questions. In particular both models imply tliat there should be a body like the board of directors to organise the relationship between share holders and management. However they differ in the role that they assign to this board. Both models are discussed in section 1I.A. Unfortunately notwithstanding the illuminating insights they offer, these models leave important questions unanswered. This issue is discussed in section 1I.B.
A. Share holder versus stake holder model The share holder model starts from the problems share holders face once management and ownership are at least partially separated, as is the case in many large firms in practice. In particular, because of its involvement with daily operations, management acquires an informational advantage relative to equity owners. This creates opportunities for the former to choose corporate policies that best serves its own goals rather than policies that are best for the firm or the share holdersl. Contrary to what is often intuitively thought shares are not a strong but rather a weak type of security. This weakness is caused by the residual character of the cash flows of equity. In particular capita1 is only repaid after all other claimants have received their due. Contrary to debt, there is neither periodic review or maturity of the contract nor any fixed interest payment. Furthermore, once owners have handed over their capita1 input they are not needed anymore to keep the production process going. This is far less the case for other factors of production like labour. However, just because of its flexible character, companies can hardly survive without equity. These special properties also make it hard to protect share capita1 from self interested actions of management by simple contracts. Hence a stronger instrument that matches the flexibility of share capital is needed.
This instrument is the board of directors and according to the share holder model the protection of equity holders is the only reason and simultaneously the explanation for the observed existence of boards. Other factors of production are 'strenger' andlor have very specific problems and can therefore be protected sufficiently by other means like contracts (e.g. debt) or special purpose governance meclianisms (e.g. collective wage ~iegotiations,work councils,...). That is, a board is a genera1 governance instrument and it is not cost efficient to use it where a special purpose governance teclinology would do. Such a loss of efficiency can be easily illustrated by the following case. Imagine for a moment that next to its share holders, some firm would also have debt holders, labour, .... represented on its board. It is not hard to imagine how in such a situation decision making could easily be hampered by continuous negotiations between different parties on the board. On top, ultimate choices would be determined by the at times haphazard way in which majorities are attached through the formation of specific coalitions. In fact it is wel1 known from the literature on voting that if preferences across classes of voters are different, even the order in which problems are placed on the agenda may strongly affect decision making. Hence, according to the share holder model, next to representatives of the management team who are needed for the information flow, one should find only representatives of the equity owners. Stake holder and share holder model only differ on the issue of board representation. In particular, the stake holder model posits that next to share holders also other parties make residual firm specific investments. In fact any one who has made such type of an input is a stake holder. Often cited examples are the effort spent on building up company specific human capital or the investment sunk in a production process tailored to the need of an important industrial client. Likewise equity such an input is also residual in nature. Hence board representation is the appropriate instrument to protect it. The proponents of this model do not offer a solution to the earlier described negotiation problems. Obviously these difficulties burden the practical application of this view. Another element that hampers the use of this model is confidentiality. In particular, some decisions like strategic choices require confidentiality for success. However, it is clear that board members may serve the class they represent best by not keeping to secrecy. For example, it is easy to imagine that such a problem may occur for labour representatives during merger talks where
plans involve a downsizing of the number of employees. Hence if a governance model is to function in practice, either it should keep close enough to the share holder model or the firm should be in a specific situation where it can circumvent the problems associated with the stake holder view.
B. Theoretica1governance rnodels offer an incomplete solution It is clear that the logic of the preceding models does not offer a complete solution to the governance problem. For, who assures that the board of directors will actually take up its tasks as it should? Experience shows that it is not enough to prescribe by law that the directors should diligently perform their duties. To guarantee board performance one could think of creating some supervisory body that supervises the board. However in turn, who will guarantee that this monitor of the board will fulfil its duties? This problem of who monitors the monitor is hard to solve. The US governance model discussed below addresses the issue by creating a system of checks and balances (or countervailing forces). In particular, if one can find a party that serves its own self interest by actively keeping the board and hence also the management on its toes, market forces will take care of the monitoring-of-the-monitor-problem.In the pre-1990 governance model of the US, this was the threat of a take-over. In the post-1990 period the countervailing force is share holder activisin. Logic and the US experience indicate that the success of a governance model heavily depends on the quality of the pressure and monitoring exercised by this countervailing force. The current critique on the quality of the Belgian governance system (discussed below) can also be regarded as an illustration of the preceding stateinent. In particular, in Belgium the role of countervailing force has traditionally been played by large dominating share holders. The main issue in the current governance debate is not that these owners have exercised insufficient pressure but rather that they have exercised too much of it. Specificaly, as indicated in the introduction, because control benefits are only reaped by the large owners, the interests of large and smal1 share holders need not coincide. The in the press heavily discussed case of the Belgian electricity provider Tractebel that is controlled by its French competitor Suez-Lyonnaise des Eaux clearly illustrates the problem at hand. Hence the in Belgium still ongoing quest for a countervailing force to keep the large owners in check in publicly quoted firms. In short, the
preceding discussion shows that because corporate governance is concerned with the top layer in the hierarchy of a firm a solution has to be found for the absence of a 'boss'. What is needed is a force that actively 'monitors the monitor'.
111. GOVERNANCE MODELS IN PRACTICE: T H E US (ANGLO-SAXON) CASE2 The main task of US boards is monitoring management in the interests of equity owners. This objective translates itself into the pursuit of the creation of value for the owners. Hence in this environment the fast rising popularity of a system like Economic Value Added (EVA) that aims at maximising share holder value, should not come as a surprise. The interests of other parties like workers, debt holders, customers ... come into the picture mainly as constraints on managerial decision making. In the US the view that share capita1 is vulnerable and should be protected is part of a long standing tradition. In contrast to Europe, where over the past century substantial efforts have been devoted to the protection of labour, in the US considerable attention has been paid to the protection of smal1 share holders. As explained in the introduction, US boards typically consist of inside directors - members of management - and outsiders - directors independent of management and usually also of share holders. The fact that outside directors often do not represent large share holders is obviously a consequence of the dispersed ownership structure of many publicly quoted firms. In practice this ownership structure has proven to entail important disadvantages. In particular, at least until the recent past, due to the free rider problem of dispersed ownership, it was not unusual that management had attracted sufficient clout to determined board membership? OObously under such circumstances one would not expect much board independence and strong monitoring. Nevertheless this latter type of activity is important. For one has observed in the US that in the absence of sufficient monitoring management slack grows as wel1 as the tendency of firms to over invest in often wasteful projects. The by now classica1 example illustrating this phenomenon is the growth of the US conglomerates during the sixties. Also the royal compensation packages that management receives - sometimes resulting in yearly incomes of over $ 100 million US - is likely associated with management clout. However as discussed be-
low the last several years the US corporate governance scene has witnessed important changes. A. US governnlzce closely adheres to the logic of organisational theoly
Traditionally US governance is based on three principles that reinforce each other: transparency, checks and balances and the radical treatment of conflicts of interest. In fact in inany respects the US governance system applies the textbook principles of organisational theory. For prime elements of a sound organisational structure are transparency, clear separation of functions and ditto behavioural rules, this in order to create checks and balances. Also structures that foster infighting and hence conflicts of interest are avoided. Until the recent past the functioning of the checks and balances proved to be the system's main weakness. Not surprisingly, the main theme of the US governance debate of the late eighties and early nineties has focused on this problem. Element 1: transparency A first t001 that helps transparency is provision of information. Consistent with the earlier mentioned long time goal of smal1 share holder protection, at least to European norms, publicly quoted firms have to provide investors with detailed information since many years. Thereby the equal treatment of al1 share holders has been a major concern to regulators. To illustrate the difference with the Belgian tradition it suffices to refer to the law prohibiting insider trading, enacted in Belgium in 1990 but inspired by long-standing regulation in the US. A second element that supports transparency is a clear set of professional behavioural norms and separation of functions with respect to the different governance parties. In particular, in the US management's task is to enact and oversee daily operations as wel1 as designing corporate strategy. The board serves as 'first-line' monitor and decides about whether or not to approve management's strategy proposals. Finally the financial markets provide financing and engage in permanent 'second 1ine'-monitoring aimed at keeping both management and board on its toes. As disciissed below, in the pre-1990 area this 'second-line' monitoring proved to be mediocre which in turn caused 'first-line' monitoring to loose quality. Because of their key role since the early nineties much attention has been devoted to develop
rules for the functioning of boards and to support the independence of independent directors. For example, currently it is considered good governance practice if the majority of the board consists of independent directors, chosen on the basis of clear professional criteria. It is equally considered good practice if, within the board, committees specialising in key issues and dominated by independent directors are created. Exainples of key issues are auditing, remuneration of top executives and hiring and firing of independent directors. Realists however understand that preceding precautions help but still cannot guarantee that the independent directors remain independent and function as the monitor of the monitor. For why should an independent director spend effort on checking the propositions of management, especially as the latter is better informed anyway? Clearly the problem of 'who-monitors-the-monitor' is hard to solve. Current US-governance practice has found a solution in strengthening checks and balances by spurring share holder activism. Element 2: checks and balances
In the US (Anglo-Saxon world), institutional investors as a group own an important fraction of the shares quoted on the national stock market(s). Their importance derives from the pension system that is based on the principle of capitalisation. Contrary to the model of repartition (as is used in Belgium) whereby those currently working pay for those currently retired, the principle of capitalisation presumes that those currently working save for their own pensions. In this way under the latter system, in a wealthy economy vast reserves are formed that need to be invested. In the US system these reserves are concentrated in the hands of pension funds managed by professional money-managers. Consequently, currently more than 50% of al1 publicly traded shares in the US are owned by institutional investors. Since the early nineties these latter owners are strongly stimulated to take up a more active role in questioning company management and boards. In fact as of the late eighties the US governance system has been steering away from hostile take-over bids (or the threat of it) as the countervailing force to keep managements and boards on their toes. During the eighties this method - which essentially involves a one shot activation of ownership (through the intermediation of the raider) - had come increasingly under attack. First of al1 this route of removing slack management is very costly as it typically involves pay-
ment of substantial premia to the target share holders. Hence it is only usable for more extreme cases of under performance. Secondly, and even more dissatisfactory, in practice it turned out that those firms that needed disciplining most, usually were not the targets of the raiders (see e.g. Franks aiid Mayer (1992)). Since the late eighties US governance has gradually been adjusting to a new model whereby permanent activation of share ownership takes a more prominent role. In particular, over the last several years US courts have increasingly upheld corporate defences against raiders. Simultaneously institutional investors have been forced to take a more active owner ship role by obliging them to vote at the general meeting of the coinpanies in which they invest. This obligation has even been extended to investments abroad. However the role of active share holder does not imply that institutional investors should strive for exercising control over firms. In fact the latter is disallowed and institutional investors may not even accept board seats. This attitude of US regulators clearly reflects the textbook view of organisational theory that the most appropriate way to create checks and balances is through separation of roles. Over the last several years, institutional investors have indeed been more active in monitoring firms. In fact it should be easier to stimulate activism wit11 institutional investors as compared to smal1 share holders. For one thing, professional money managers should be able to do a better job at monitoring than non-professionals-small-owners. Furthermore the earlier mentioned free rider problem can more easily be overcome as the ownership of institutional investors is large enough for value gains Erom activism to compensate for the monitoring costs. In practice many institutional investors have actually changed their behaviour. Instead of applying the traditional wal1 street rule (= sell if you do not like the management) they have moved towards taking a more active stance. Some have even become renowed for their activism. For example, California Public Employees Retirement System (CALPERS) and the California State Teachers Retirement System (CALSTRS) are reputed pension funds that publish black lists of firms with lackluster performance. Often share prices already improve immediately upon publication as the stock market anticipates that, under the pressure of these funds' monitoring actions, positive changes wil1 take place within the firm. Notwithstanding al1 the media attention, empirica1 evidence indicates only a limited long term direct impact of institutional activism however. In particular, the data
show little evidence of improvement in long term mal-ket performance after the targeting, and if a change in the real activities of the firm is perceived, it is hard to establish the causal relationship between activisin and the change (see e.g. Gillan and Starks (1998)). Next to institutionals, also some share holders have become more active under the influence of bulldog-firms. The latter are specialised lawyer companies that search for opportunities to initiate a court case and then attempt to receive a mandate from share holders to sue management andlor the board. Obviously, these bulldogs function as a mechanism that overcomes the free rider problem which is a plus. However one may expect its monitoring impact to be of lesser quality than that of institutionals. In fact, bulldogs may create a problem of their own because, typically these firms are paid only if a court case is won. Hence, these firms do not aim at separating out good from mediocre corporate decision making. Rather they are interested in finding decisions that can be represented in such a way that a possibly winning court case can be constructed. The averse effects of this type of monitoring can be illustrated by the problems in the field of medica1 treatment. In particular, it is not unusual in the US that doctors refuse to treat badly i11 patients because they anticipate to be sued if the treatment does not lead to a cure. Element 3: preventing conflicts of interest between share holders Eikewise the protection of smal1 owners, the reduction of conflicts of interest between share holders has been a long time goal in the US system. Already since the thirties, the legal and fiscal environment is hostile to the forination of large holding groups with publicly quoted subsidiaries. The tax treatment of complex holding structures is unfriendly, and also the antitrust regulation has hampered the formation of complicated ownership structures that often are so importailt in European countries. Next to reducing potential conflicts of interest, strict rules have been laid down for the handling of remaining problems. For example, the fact that al1 relationships between a publicly quoted subsidiary and parent firm have to be at-arm's-length (i.e. be commensurate with market conditions) has a long standing tradition in practice. By contrast, the at-arm's length-tradition in Europe is still evolving. Also, in the US, in order to obtain a quotation on the stock exchange, a company has to prove that in case conflicts of interest between share holders could occur, mechanisms are in place to
control these problems. Certaiiily in the US governance system of today, avoidance of conflicts of interest is important because of the role of monitor of the monitor played by some classes of share holders. Clearly this monitoting activity may qi~icklyloose quality if it gets polluted by class-specific goals. B. No solution is perject As mentioned before, solving the problem of who monitors the monitor is extremely hard and it is unlikely that a perfect solution exists. Clearly the choice of the US is driven by the opportunities available in its financial mass markets of today. Notwithstanding its logic, the system nevertheless has inherent weaknesses. First of all, the risk of an overly powerful management continues to loom around the corner. Secondly, asymmetries in information that are necessarily present in mass markets and the strong dependence on the quality of the judgement of outsiders may force decision makers to opt for wel1 accepted recipes even if a less traditional solution would be more appropriate. Therefore successful communication with the market has taken up a key role and firms have been adjusting to cope. To quote Ettorre (1996), 'Fifteeii years ago, the C E 0 and C F 0 did not know major share holders and really didn't care. CEO'S are now more accessible to money managers'. A third and closely related problem concerns the often heard complaint about the fact that adherence on the judgement of outsiders would force companies to realise profits in the short run. Especially the mandatory reporting of quarterly profits attracts much criticism. It goes without saying that this latter obligation forces management to spend (too?) much effort on maintaining a smooth profit flow. Doubtless this wil1 sometimes also lead to situations where, at least from the perspective of their inside knowledge, management feels obliged to take sub-optima1 decisions. However the positive impact of the monitoring by active share holders should not be under rated: as discussed aboves, an active 'monitor of the monitor' is crucial in maintaining the quality of the governance system in publicly quoted firms.
IV. GOVERNANCE MODELS IN PRACTICE: THE BELGIAN CASE As indicated in the introduction, the governance situation in Belgium as wel1 as in most other continental European countries is quite different from the one in the Anglo-Saxon world. The most striking difference probably is the concentration of ownership. Although also in the US publicly quoted companies may have large or even majority owners, overall the ownership is much more dispersed there. Notwithstanding the overall picture of stronger concentration of ownership, still important differences between European countries may exist. In particular comparative evidence indicates that the percentage of publicly quoted firms controlled by a large share holder or syndicate of large equity holders is higher in Belgium than in its neighbouring countries (Wymeersch (1997)). Consequently, large share holders typically dominate Belgian boards, and hence also the process of hiring and firing management. Furthermore, as they invest in following up the firm, these owners strengthen their influence even more relative to the passive smal1 owners. In this way - at least as compared to the US where overall there is a clear seperation of rules - in Belgian practice usually there is not such a clear distinction between the roles the different governance actors actually play. Bbviously the main advantage of this method of organising top decision making consists of the close attention large share holders pay to what happens in the company. Most importantly, although it is different froin the US (Anglo-Saxon) method, the Belgian way of activatiiig share ownership also offers a solution for the difficult problem of who 'monitors the monitor'. However this method of handling the latter issue also raises two important questions. First, are conflicts of interest between share holders resolved (i.e. this is a question about the quality of the monitoring by the monitor of the monitor)? Second, is this type of governance optima1 for al1 firms at al1 times? A. Belgian governance and conflicts of interest Notwithstanding multiple attempts to resolve conflicts of interest between classes of share holders, this problem still remains at the heart of the governance discussion in Belgium. An important step in tackling the issue was taken in 1995. Since then corporate law requires that any transaction between a parent and its publicly quoted subsidiary
should be evaluated in a report drawn up by an expert and three directors that are independent of both management and share holders (Ralet (1996)). Although such a measure is useful, still it cannot prevent many types of conflicts. A parent that consistently steers business opportunities to other firms of the group is only one possible example. Another one is the case where the company is forced to adjust its strategy to fit its self financing capacity. Furthermore, as discussed in section 2, without checks and balances it is difficult to keep the independent directors independent. Whereas in the US independence with respect to management is the issue, in this case the problem concerns maintaining independence from the large share holders, especially as these owners typically determine board seats. Recently severa1 organisations have come forward with propositions for improvement. The Commissie voor het Bank en Financiewezen suggests rules for more transparency. The commissions Cardon and Santens as wel1 as the VBO focus on the board and propose rules that should lead to improvements in the quality of board functioning. A common feature of the latter reports is that they contain rules aimed at making boards more autonoinous relative to large owners. For example, a typical suggestion is that boards should have independent directors; als0 the creation of specialised committees with an important role for the latter board members is suggested as good governance practice. Clearly these propositions are inspired by US-style governance, but certainly do not go al1 the way. This is understandable. For the moment experience with a governance system in which independent directors play an important role is lacking. Furthermore as indicated earlier, the US (Anglo-Saxon) evidence shows that the most difficult problem with this type of directors is to keep them independent. Without checks and balances it wil1 be difficult to avoid that over time they become dominated by the large owners. On the other hand, with the more radical solution that requires that independent directors fill the majority of board seats, the problem of dominance by management looms around the corner. Finally, even a wel1 functioning board still cannot resolve al1 conflicts of interest. For example, one could imagine that at the general meeting, for purposes of serving the interests of the group, a parent firm uses its voting power to block the propositions from the board of a publicly quoted subsidiary.
B. Is a uniform governance model good? A striking feature of current Belgian governance is the almost uniform application of the governance model with dominating share holders. As indicated above, concentrated ownership is the typical method by which in continental European countries (certain classes of) share holders in publicly quoted firms are activated. Preceding discussion certainly does not imply that this type of activation is inferior to tlie inarket based activation in the US. What is important is the motivation underlying the choice for concentrated share holder ship. Opportunities to derive important control benefits based on conflicts of interests between owners is a negative reason for concentrated ownership; another negative reason is under developed capital markets. By contrast, a much better motive concerns reaping a sufficient return on managing the firm wel1 (i.e. the case of the manager-owner) or reaping a return on the investment of monitoring management. In fact some firms prove to be better off with concentrated ownership. In particular, numerous US companies with weak growth potential but strong cash flow have realised dramatic improvements in profitability after share holders from the public at large were bought out and hence a more concentrated ownership structure was realised. However after major operational restructuring has taken place, often these firms are taken public and may end up again with (relatively) dispersed ownership. Furthermore although dispersed ownership is an important feature of US governance, large block holders are not unknown in the US. In fact Shleifer and Vishny (1986) mention that some 30% of the Fortune top 500 companies have large owners. However it should be understood that in the US a block of 5 to 10% of the shares represents a very important ownership position. By contrast in Belgium, a block of this size is still considered to be relatively small, and unless a coaliiion can be formed with the dominating share holders, it may not yield influence over firm decision making. Preceding USevidence contains a very important suggestion concerning ownership however. In particular, it indicates that it is highly unlikely that in a dynamic economy al1 publicly quoted firms are at al1 times best off with concentrated ownership. The observation that in a particular country al1 firms adhere to this type of owner ship structure should trigger questions about the underlying motives. Although the argument of an undeveloped capital market may have had relevance in the past, it is becoming quickly less satisfactory as an explanation for the
currently observed ownership structures in Belgium. Furthermore, and this contains a further negative indication for the Belgian situation, international evidence reported in La Porta, Lopez-de-Silanes, Shleifer and Vishny (1997) indicates that the degree to which publicly quoted firins adhere to concentrated ownership is positively correlated with unresolved conflicts of interest. V. US (ANGLO-SAXON) GOVERNANCE IS A SUCCESSFUL EXPORT PRODUCT: A RATIONALE
Although the Anglo-Saxon type of governance is dominant only in a small fraction of the financial markets of the world, i.e. in the AngloSaxon capita1 markets (other countries tend to base their governance system on concentrated ownership), this governance model has nevertheless been extremely influential in European governance discussions. In fact for several reasons one may expect that in the future its impact on Europe will continue to rise. In particular, within Euroland, stock markets are quickly integrating so that a financial mass market is likely to emerge within the next few years. In this mass market institutional investors will be important actors and companies will have to adjust to these investors' demands. Thereby the latter typically consider a firm's adherence to the Anglo-Saxon governance rules as a stamp of quality. As in financial mass markets quality labels are important4, and as Anglo-Saxon rules are suited to be used in mass markets and as on top no alternative quality label is available, one may expect increasing pressure on European countries to move closer towards the Anglo-Saxon model. Other developments are als0 likely to force Belgian (European) publicly quoted firms will have to pay more attention to the preferences of the group of non dominant owners. In the future, in particular, the lure of the historically high stock price levels lias led an increasing number of firms to undertake an initia1 public offering or a seasoned offering during the late nineties. In fact, until the recent past very few Belgian (continental European) firms ever considered the idea of getting a public quotation. Moreover once quoted, these companies rarely tapped the stock market for fresh equity capital. This together with a limited free float and hence small liquidity of the market in firms' shares, contributed to the fact that limited attention was paid to the stock market in corporate decision making. Obviously, when raising cash from the public market becomes more important,
this necessarily implies that more attention has to be paid to the latter. Another factor that causes a rise in companies tapping the financial resources of the market are new growth opportunities. Whereas in the past most Belgian firms were able to adjust their growth plans to fit self financing, for an increasing number of companies this has become impossible. In particular, contrary to the more traditional businesses where the size of the market is often geographically limited, the high-tech sector is concerned wit11 products whose natura1 home market reaches far beyond the borders of the country where the firm is located. Furthermore developments in that area often imply that if a company is to prosper it needs to attain quickly a leading position world wide or at least continent wide. Finally, in several business sectors, significant increases in scale through mergers and acquisitions are taking place. Again, the latter type of transactions requires additional financing. Next to the increased appetite for additional financing, also overall changes in the environment are already today causing (some) companies to change their attitude towards the stock market. First and foremost, with the Euro, the home market for Belgian investors will continue to broaden to encompass Euroland. This implies that the traditional home-bias (i.e. unvestors' preference for securities from the home country) wil1 weaken. One may expect that this tendency to internationalise portfolios will be enhanced by the rising importance of institutional investors. In turn this will force firms to compete internationally for investors' money. In fact the example of Sweden shows that the pressure of institutional investors may quickly force companies to adjust their governance system to a more inarket orientedview. Even ownership structures have changed over a time span of a few years after Anglo-Saxon pension funds increased their investments in Swedish firms. In particular, Rydqvist and Odegaard (1997) report that the percentage of publicly quoted firms with more or less widely dispersed ownership (i.e. no one owns more than 25% of the shares) increased from 5% to 24% in the period 1991-1996, and this primarily due to a different ownership structure of newly quoted firms. Simultaneously the traditional premium for shares with multiple voting rights vis-à-vis those with single voting rights dwindled to almost zero. A second change in the environment that could deliver an important contribution to firms' preoccupation with the stock market is the increasing interest in and success of stock option plans (in Belgium and Europe). International competition for top talent places especially Eu-
ropean liigh-tecli firms under an increasing pressure to match US (Anglo-Saxon) payment conditions. Whereas in the past the legal and fiscal system has been hostile to it, since the mid nineties several European countries (among which Belgium) have been adjusting rules in favour of this type of remuneration, whereas others are working on proposals in that direction. It is important to note that preceding discussion does not imply that concentrated ownership wil1 disappear in Belgium (Europe). For the problem of European governance is not that large ownership would be a bad thing. In fact, as previously indicated, large share holders inay be a plus for the governance system as they help to resolve the problem of 'who monitors the monitor'. Rather the main issue in the European discussion concerns the fact that conflicts of interest have not been sufficiently addressed. And, as meiltioned in the previous section, also in the Anglo-Saxon world numerous prospering publicly quoted firms have a large block holder. VI. WILL CORPORATE GOVERNANCE TRAWSFORM T H E BELGIAN (EUROPEAN) FINANCIAL SECTOR? Notwithstanding the fact that European corporate governance is likely to undergo major adjustments in the next few years, one may not expect these changes to transform the Belgian (European) financial sector. Rather the evolution in governance is part of a major process of transformation of financial markets that is generally expected to deeply affect banking firms. Many articles have already been devoted to the impact of future trends on banking firms5.A discussion of these expected changes obviously falls outside the scope of the present paper. Nevertheless a few comments on corporate governance and the transformation of European financial markets are in order here. As discussed in section V, future trends are likely to trigger important adjustments in corporate governance. However, conversely, the earlier mentioned evidence from La Porta, Lopez-deSilanes, Shleifer and Vishny (1997) also implies that a more market oriented governance model as the one of the US better supports a financial mass market in equity and bonds. Put differently, US-type governance would support desintermediation. For banks this obviously has a negative impact because lending business may be lost. Simultaneously however, desintermediation and a governance model that is more open to outsiders creates more invest-
ment banking opportunities for financial intermediaries. Furthermore if tlie more market oriented governance view spills over into the sector of non quoted firms, tlie venture capita1 subsidiaries of banks may also profit from a rise in activity. On tlie negative side, traditional saving products offered by banks will likely loose interest from the investing public. On the positive side however the changes in the financial markets will create opportunities for new products and more fee business from money-management. Overall at present it is still to early to make predictions about whether advantages will outweigh disadvantages. Furthermore such a prediction exercise is hainpered by the fact that not only European banlcing models differ from country to country, but also that within countries the market position of intermediaires and hence their likely reaction to changes in the environment may differ also. VII. CORPORATE GOVERNANCE LOGIC AWD SPECIFIC REGULATION ISSUES: TWO BELGIAN EXAMPLES In the introduction the question was asked whether ideas from corporate governance could illuminate or at least offer additional perspectives with respect to specific regulatory issues6. This section illustrates that, at least for the problem of bank ownership of share capital in other firms, this is indeed the case. In particular, subsequent subsections V1I.A. en V1I.B. offer some comments on the issue of permanent ownership of large blocks in publicly quoted firms and the issue of transitory ownership of shares in non quoted companies from the governance perspective. A. Large 'permanent" owner ship bloclzs in publicly quoted firms Ever since the abolishment of the mixed bank in Belgium in 1935 and the ensuing split of its business into the activity of commercial banks on the one hand and the business of holding firms on the other, at regular intervals tliere has been an upshot in the debate about whether or not commercial banks should be allowed to take large perinanent positions in the ownership of publicly quoted firms. Although over time the original ban on share ownership has been weakened to allow banks to permanently own equity in financial subsidiaries and take transitory ownership positions in the framework of investment banking activities, the genera1 principle still remains that commercial
banking and the business of holding firms should be kept separate. In the past the usual arguments in favour of lifting the ban concerned diversification and provision of additional risk capital. In particular, time and again it was argued that publicly quoted Belgian firms had difficulty attracting investors to provide them with sufficient equity financing. Banks could help solving the problem by investing in shares and hence channel more savings towards equity capital. This argument was often complemented with the reasoning that although shares are ris@ and hence equity investments could increase the risk profile of the bank, such an increase should not be over estimated. Specifically, this type of activity would allow banks more diversification possibilities and hence give them more opportunities to generate income from multiple sources. However today the ??? are likely to outweigh the ???. First there is the Longstanding argument rooted in the experience from the economic crises of the thirties where losses on equity positions brought banks int0 difficulties. Furthermore, with the development of Belgian capital markets, and in view of the impending integration into a large Euro-market, at present as wel1 as in coming years, risk capita1 is (likely to be) sufficiently available. Finally the arg u m e n t ~from corporate governance given below indicate that in today's capita1 markets the disadvantages of banks as large block holders probably outweigh the advantages. According to the academic literature, the main advantages of using banks as large share holders-monitors are threefold. First there is the opportunity to reduce information asymmetries thereby benefiting both banks and borrowing companies. Second, some conflicts of interest may be mitigated when banks also become share holder. Third, large share holders are strong monitors. The hypothesis that information flows between banks and borrowing firms are favourably affected if a banker can simultaneously play the role of lender and share holder is based upon the observation that both lending and monitoring require inside information. Clearly if the banker-large share holder is also represented on the board, it has two channels to receive complementary information. This reduction in information asymmetries is hypothesised to result in many beneficia1 effects. For example, it is argued that because the banker is better informed, it may be less reluctant to extend loans because it understands that the management lacks the necessary informational advantage to f001 the banker. Furthermore since the financial intermediary is both share holder and lender, the conflicts of interest between the providers of those two
kinds of financing is reduced (see for example Miilbert (1998)). However, when the firm is publicly quoted these advantages also imply conflict~of interest between tlie banker and the small share holders. For the intermediary's motives can be regarded as a portfolio of share holder and lender interests in which the weights of the different views vary according to the relative benefits of its ownership stake and lending activities. In fact, the basic argumentation concerning the rationale for a board from section 11 implies that bringing debt interests onto the board causes governance inefficiencies if taylor-made protective mechanisms for debt are available. In particular to protect itself against unanticipated swings in corporate policy tnat could impede its position, a debt holder can use protective covenants. Furthermore the whole idea of having publicly quoted companies is that there is enough information available to allow share holders from the public to make an assessment of the firm's situation so that the stockprice reflects future opportunities in a reasonable way. One would expect that this information also suffices to make reasonable assessments about the quality of a loan. Hence the ultimate remaining advantage of the mixed banking model is the one of the strong monitoring by a large share holder. However from the previous discussion it is obvious that banker-lenders are not the ideal large owners because of conflict~of interest. Furthermore because the business ties between int e r m e d i a ~and firm are close, it is very difficult to avoid these problems. As indicated earlier, the governance discussion in Belgium (and many other European countries) essentially centres on the impact of large share holders on the quality of firm governance. What can be learned from the US governance system of today is that avoidance of conflicts of interest is important once certain classes of share holders fulfil the role of monitor of the monitor. As discussed earlier, the quality of the governance system heavily depends on the quality of monitoring exercised by the latter, and class specific goals may quickly undo beneficia1 effects. Finally, next to academic arguments, also more practica1 oriented reasons in favour of the mixed banking model have been offered. In particular it has sometimes been argued that the banker-share holder could provide firms with financial and fiscal know-how as wel1 as access to business networks. It is unlikely that these arguments have practical relevance though. For, although this may be true for small companies, it is doubtful that publicly quoted firms are insufficiently
professionalised so as to be in need of permanent support in these areas. Mulbert (1998) discusses empirica1 evidence on Germany where share ownership of banks in publicly quoted firms is important and finds that the results are inconclusive about the benefits of bank block holders. In particular, some studies find that banks do a good job at monitoring industrial firms, while other research indicates that bank interference leads to no better or even inferior results. Especially for later periods, the majority of the studies find average or even inferior performance of banks as large block holders as compared to non bank large block holders. In particular Gorton and Schmid (1996) considered two samples: one for 1974 and another one for 1985. For the first period, company performance was positively related to bank holdings, while for the second period no relationship was found. For a sample comprising the period 1988-1992 Nibler (1995) confirms the results from Gorton and Schmid's 1985 sample and reports that banks do not perform their monitoring task any better than other large share holders. Wenger and Kaserer (1998) find for the period 1974-1993 even a negative relationship between block holding by German banks and company performance. Along the same lines Böhmer (1997) reports that in take-overs bidder share holders realise a smaller return if banks are important owners of the bidding firm. B. Transitoly financing of n o n quoted firms by specialised subsidiaries (= captive venture firms)
From the perspective of governance arguments, the case of banks temporarily owning blocks in non quoted firins is quite different. For the latter type of companies the earlier arguments about information asymmetries and effective support in terms of financial, fiscal, organisational know how and access to networks are much more likely t o be of practica1 significante. Furthermore as these firms are non quoted, the problem of conflicts of interest with smal1 share holders does not exist. What remains however are possible conflicts of interest embedded in risk taking by the bank vis-à-vis its depositors and its other debt holders. Current regulation minimises such conflicts as banks are required to cover investments in shares in full by bank equity. F ~ i r thermore as regulation does not permit other equity positions than those in financial subsidiaries and transitoiy ownership in the framework of investment banking activities, Belgian banks are forced to con-
centrate venture capita1 activity in specialised subsidiaries. This latter policy adds to transparency and obviously is a plus from the perspective of good governance. VIII. CONCLUSION This paper discusses the logic of the share holder and the stake holder model for corporate governance and the way these theoretica1 ideas are applied in the actual governance system of the US (Anglo-Saxon) and Belgium (Continental Europe). Several major conclusions follow. First, a good governance solution for publicly quoted firms requires that the problem of 'who monitors the monitor' is addressed. In turn this implies that to be successful, actual governance systems need the integrated use of several tools. Second, because of the importance of quality labels in financial mass markets, and the fact that so far Europe has not been able to build a consensus on a workable alternative, the US (Anglo-Saxon) view of governance is likely to gain influence in European stock markets. Finally, when corporate governance arguments are applied to the problem of share ownership by banks within the Belgian context, it is seen that governance models add useful perspectives to this regulatory issu. NOTES 1. This agency problem was í'irst aiialysed by Coase (1937) and by Jenscn en Meckling (1976). 2. Sectio11I11 and IV follow the discussion i11 Van Hulle (1999). 3. The free rider problem refers to the situation in which 110 individual small share holder tindertakes costly inoiiitoring because thc gain this owner reaps o11 hisllier owncrsliip position is too small to ofí'set the costs. Furthermorc al1 share liolders participate i11the gains and tliercfore every iiidividual owncr waits for sonicbody clse to initiale thc monitoring aiid free ride on the lalter's moiiitoririg investment. 4. Two proiniiierit problcms in mass markets are iniormation costs and information asymmetries. Quality labels help to reduce tliese problems and heilcc help thc mass market to reniain viahle. A prominent example of tlie usc of such labels is deht rating. 5. Tlie tcrms 'bank', 'fiiiancial firm' and 'fiiiancial intermediary' wil1 be used intercliangeably. 6. For more gcneral discussion about tlie intcraction of governaiicc and bank reg~ilation, see Van Hulle (1999).
Bölimcr E,, 1997, Iiidustry Groups, Bank Coiitrol, aild Large Share I-Iolders: ai1 Aiinlysis of German Tal
Commissie Santens en VKW, 1996. Commissie voor liet Bank en Financiewezen, 1998, D e aanbevelingen van de Commissic voor het Bank-en Financiewezen aan de Belgische veiinootschappen i.v.in. hun iriformatie over de wijze waarop zij hun bestuur e11beleid organiseren, 6. Dacms H., 1998, D c paradox van het Belgisch kapitalisme, (Lannoo, Tielt). De Keulerieer E., 1994, Levert een beursnotering in België voldoende voordelen op?,Batiken Rilnt~cieivezen,9-10, 531-535. Ettore B., 1996, Whcn Patience is Corporate Virtue, nilrnlagement Review 85,28-32. Franks J . and C. Mayer, 1992, Hostile Takeovers and the Correction of Managcrial Failure, discussion paper, (London Business School). Flamee M. and D. Meulcmans, 1997, Codex financieel recht 1997.1998, (Die Kcure, Brugge). Gillan S. and L. Starlts, 1998, A Survey of Share Holder Aclivism: Motivation and Empirical Evidence, Coizteiizpomiy Finunce Digt>.rt,autlimn, 2, 3, 10-34. Gorton G. and F. Schmid, 1996, Universal Banking and the Performance of German Firms. NBER working paper 5453. Jensen M., 1989, The Eclipse of the Public Corporation, Naivard Business Review, September-October, 61-74. Jensen M. and W. Meclding, 1976, Theory of the Firni: Managerial Behavior, Agency Costs and Ownership Structure, .Tourria/ of Financial Ecolzoi71ic~,305-360. La Porta R., F. Lopez-de-Silanes, A. Shleifer and R. Vishny, 1997, Legal Deterininants of External Finance, Jo~lnzalof Finui~ce,July, 52, 3, 1131-1150. Mï~lbertP., 1998, Bank Equity Holdings in Non-Financial Firms and Corporate Goveriiance, in IC. I-Iopt, H. Kanda, M. Roe, E. Wymeersch, S. Prigge, eds., Comparative Corporate Governance. The State of the Art and Emerging Research, (Clarendon Press, Oxford), 445 - 564. Nibler M., 1995, Bank Control and Corporale Governance in Germany: the Evidence, Research paper no 48, (Faculty of Economics and Polities, University of Cambridge, U.K.). Ralet O., 1996, Responsabilités des dirigeants de sociétés, (Larcier; Bruxelles). Rydqvist K. and B. Odegaard, 1997, The Valuation of Corporate Voting Rights and Portfolio Diversification, paper, (Norwegian School of Management). Shleifer A. and R. Vishny, 1997, A Survey of Corporate Governance, Jo~lmalof Finunce, June, 737-783. Shleifer A. and R. Vishny, 1986, Large Shareholders and Corporate Control, Journal ofPoliticnl Econonty, June, 461-488. Vali I-Iulle C., 1999, Bank Regulation and Bank Corporate Governance: Flip Sides of the Same Coin? A Note, mimeo (Department of Applied Economics - KULeuven, Leuven). Van Hulle C., 1999, Belgische Corporate Governance in (snelle?) beweging, in Bedrijfsbeheer en Taalbedrijf, Jubileumboek 30 jaar VLEKHO; Van Hulle C., 1998, Is het systeem van corporate governance belangrijk? Op zoek naar de impact van verschillen in modellen, in R. De Bondt aiid R. Veugelen, eds., Informatie en kennis in de economie, Drieëntwintigste Vlaams wetenschappelijk economisch congres, (Universitaire Pers Leuven), 287-355. Verbond van Belgische Ondernemingen, 1998, Corporate Governance. Aanbevelingen van het VBO, 12. Wenger E. and C. Kaserer, 1998, The Gerinan System of Corporate Governance: a Model Which Should Not Be Imitated, in Black S. and M. Moersch, eds., Competition and Convergence: the German and Anglo-American Models, (Washington D.C.), 40-81. Williamson O., 1984, Corporate Governance, Yale Luw Joumal 93, 1197-1230. Williamson O., 1985, The Economic Institulions of Capitalism, (The Free Press, NewYork). Wilmots H., 1999, Corporale Governance vanuit het hart van de interne organisatie, Financieel Economische Tijd, 2 april, 7. Wymeersch E,, 1997, Corporate Governance: waarheen?, Bank- en Financiewezen, November, 9, 602-606.
Tijdschrift voor Economie en Management Vol. XLIV, 3,1999
naditionele versus optie-waarderingsmethodes voor groeibedrijven: er zijn geen mirakeloplossingen P. SERCU' en C. VAN HULLE*"
I. INLEIDING De laatste jaren is er - vooral in de academische wereld - een snel groeiende interesse voor het toepassen van optiemethoden bij het waarderen van ondernemingen. Volgens de proponenten zou het gebruik van deze techniek de kwaliteit van de waardebepaling van heel wat bedrijven, en in het bijzonder deze van groei-ondernemingen, ten goede komen. Soms wordt hierbij zelfs geargumenteerd dat het toepassen van het 'verdisconteerde dividenden'-model (DD), sowieso niet opgaat voor groeibedrijven. Bovendien zouden zakenbanken, die de traditionele 'verdisconteerde liasstromen'-benadering (DCF) plegen te gebruiken, de waarde miskennen die vervat zit in de mogelijkheid om, afhankelijk van de binnenkomende informatie, toekomstige investeringsplannen aan te passen. Zo heeft men opgemerkt dat in de jaren na beursintroductie de aandelen van groeibedrijven met meerdere honderden procenten kunnen toenemen. Sommige beurswaarnemers schrijven dit dan ook toe aan de onderwaardering bij beursintroductie, en een aantal academici verklaren op hun beurt deze onderwaardering door de daarnet vermelde tekortkoming in de gehanteerde waarderingsmethodes. In deze bijdrage pleiten wij voor enige voorzichtigheid rond deze stellingen. We onderschrijven wél de stellingen dat DCF als waarderingsmethode aan te bevelen is boven DD, * Graduate School of Business Studies, K.U.Leuven, Leuven.
'::"Departement Toegepaste Economische Wetenschappen, K.U.Leuven, Leuven.
323
dat flexibiliteit vaak over het hoofd gezien wordt, en dat optietechnieken conceptueel superieur zijn aan DCE Toch lijken ons hierbij enkele randbemerkingen nuttig: Het lijkt ons onwaarschijnlijk dat de hausses die op menige beurs aan de aandelen van heel wat nieuw geïntroduceerde groeibedrijven te beurt vielen, te wijten zijn aan verschillen in de waarderingsmethodes gehanteerd door emittent-plus-zakenbank en secundaire beleggers. In principe zijn DD en DCF equivalent, maar met DCF maakt men minder kansen op fouten, en dat geldt zowel voor groeibedrijven als voor gevestigde waarden. Een flink stuk van de door een groeiend aantal academici aangehaalde bezwaren tegen DCF, in vergelijking met optiemethodes, zijn niet inherent aan DCF, maar verwijzen eerder naar courante fouten in de toepassing van DCE e De werkelijke bijdrage van optiemethodes ligt eerder in het modelleren van afgeleide risico's; en de praktische toepassing ligt bezaaid met voetijzers en schietgeweren. Optieprijsmethodes leveren diis geen mirakeloplossing. Elk van die punten komt hieronder aan bod. 11. D E BEURSGANG: INITIELE VERSUS LATERE RETURNS EN HUN OORZAKEN Op menige beurs kan men - vooral sinds het midden van de jaren 90
- vaststellen dat tijdens de eerste jaren na de introductie de aandelenkoers van groeibedrijven al eens verveelvoudigt. Ook in België heeft zich dergelijk fenomeen voorgedaan. Zoals in de inleiding aangehaald wordt soms geopperd dat zakenbankiers en emittent de bal zwaar verkeerd slaan bij de initiële waardering van dergelijke bedrijven. Om dergelijke uitspraak te evalueren lijkt het ons nuttig een onderscheid te maken tussen enerzijds het initiële rendement (de appreciatie tussen de introductieprijs en de eerste evenwichtskoers op de beurs), en anderzijds het rendement in de jaren na de eerste notering. Initiële koersstijgingen zijn vaak substantieel (gemiddeld enkele tientallen percenten), maar koerswinsten in de grootte-orde van honderd(en) percenten, Als die zich voordoen, worden gerealizeerd in de jaren ná de introductie. Laat ons eerst het initiële rendement bekijken. De koersstijging tussen introductie en eerste notering(en) is een internationaal fenomeen:
studies in diverse landen leveren gemiddelden op van 5 tot 50% (Ritter (1984), Van Hulle, Imam, en Casselman (1993)). Maar die studies wijzen er tevens op dat dergelijke positieve opbrengsten vooral geconcentreerd zijn in euforische 'hot issue periods'. Zouden de beleggers, bij het vaststellen van de eerste werkelijke notering, dan alleen optieprijsmethodes toepassen in euforische periodes? Er zijn overtuigender verklaringen voor het initiële rendement. De kostenvan een mislukte introductie, zowel rechtstreekse kosten als reputatie-effekten, zijn hoog. Ook de voordelen van het vestigen van een succes-aureool bij potentiële klanten en kapitaalverschaffers zijn erg belangrijk. Dus gaan emittent en zakenbank rationeel en bewust tegen een veilige marge beneden de geschatte waarde uitgeven (voor een overzicht van gelijkaardige argumenten, zie bij voorbeeld Vandemaele (1997)). En bij groeibedrijven, waar de onzekerheid over de eerste effectieve waardering door de markt groot is, wordt de veiligheidsmarge bij de inschrijvingsprijs overeenkomstig groter. Initiële ondenvaardering heeft dus vermoedelijk andere oorzaken dan onbedoelde onderschatting door zakenbank en emittent. Is het rendement nà de eerste notering dan wél aan superieure waarderingkunst bij de secundaire belegger toe te schrijven? In tegenstelling tot het gemiddeld positieve initiële rendement is het fenomeen van hoge latere returns niét het standaardpatroon. Internationale evidentie wijst erop dat tot in het begin van de jaren 90 nieuwkomers op de beurs achteraf eerder ondermaats presteerden. Ook België is niet aan dit fenomeen ontsnapt (Imam en van Hulle (1998))'. Bovendien is het naast elkaar voorkomen van zeer goede en zeer slechte beursprestaties in de jaren na introductie in het geheel niet verwonderlijk, zeker niet bij groeibedrijven.Immers, de investeringenvan dergelijke ondernemingen kunnen net zo goed tegenvallen - kijk bijvoorbeeld gewoon naar de hemelhoog geprezen (en geprijsde) biotech-bedrijven van vijf en tien jaar geleden, en de middelmatige performance op middellange termijn van tijdens de jaren 70 en 80 geïntroduceerde ondernemingen in de VS. We concluderen dat de initiële return waarschijnlijk niet met onbewuste onderwaardering te maken heeft, en dat er weinig sporen zijn van onderwaardering in de langere periode na de eerste notering. Daarom lijkt het onwaarschijnlijk dat er systematische verschillen zouden zijn tussen de waarderingsmethoden van emittent en zakenbank enerzijds, en de gewone belegger anderzijds. Een dergelijke visie zou overigens geïmpliceerd hebben dat, in tegenstelling tot emittent en
zakenbank, Jan met de pet wél weet hoe men bedrijven korrekt hoort te waarderen.
111. DIVIDENDEN OF KASSTROMEN? De laatste jaren is de voorkeur van vele zakenbankiers inzake gehanteerde waarderingsmethode verschoven van verdisconteerde dividenden (DD) naar verdisconteerde kasstroom (DCF), zeker in het geval van groeibedrijven. Immers, veelal (en terecht) wordt gesteld dat de meeste jonge ondernemingen geen dividenden uitkeren omdat zij investeren; enkel bedrijven die al enkele jaren voldoende kasstromen generen en een veeleer stabiele en voorspelbare dividendpolitiek volgen, kunnen met DD worden gewaardeerd. De lezer zou hieruit ten onrechte kunnen concluderen dat DD en DCF verschillende benaderingen zijn. Onze stelling is dat (a) in theorie beide equivalent zijn, en (b) in de praktijk DCF over het algemeen aan te bevelen is, of het nu gaat om groeibedrijven of niet. In principe is DD het logische vertrekpunt: de belegger betaalt orndat (en in de mate dat) hij geld terugvenvacht. Het is echter duidelijk dat dividenden (inclusief terugkoop van aandelen, inschrijvingsrechten, enz.) inderdaad moeilijk te voorspellen zijn bij groeibedrijven. De onzekerheid over de uitkeringen reflecteert onzekerheid over twee onderliggende vragen: (a) hoe groot is, elk jaar, de beschikbare kasstroom, en (b) wordt die kasstroom uitgekeerd dan wel voor onbepaalde tijd gereserveerd? De laatste bron van onzekerheid is van secundair belang. In perfecte markten is het moment van uitbetaling zelfs totaal irrelevant, zoals Miller en Modigliani al in 1961 aantoonden: als een bedrijf de uitbetaling uitstelt zonder de investerings- en operationele politiek te wijzigen, wordt er gewoon later méér uitgekeerd - inclusief de ondertussen verdiende rente, en dus zonder enig effect op de huidige waarde van het aandeel. Of anders geargumenteerd: als een bedrijf honderd miljoen uitkeert, zakt zijn beurswaarde op het ogenblik van de uitkering ook met honderd miljoen, zodat de totale rijkdom van de aandeelhouder (aandelen plus kasgeld uit uitgekeerde dividenden samen) niet verandert. Die logica is meteen ook de basis van de DCF-procedure, waarbij de kasstromen zélf gewaardeerd worden eerder dan de werkelijk verwachte uitbetalingen. Bij DFC doet men dus alsof de hele kasstroom
onmiddellijk via dividenden enlof wederinkopen uitbetaald wordt, aannemend dat, zoals in perfecte markten, de mogelijkheid tot uitstel geen bijkomende impact heeft op de waarde. In de praktijk zijn kapitaalmarkten weliswaar allesbehalve perfect, maar de voorkeur die thans vele zakenbankiers uitspreken voor DFC in plaats van DD betekent dat zij (net zoals ondergetekenden, overigens) bereid zijn te leven met het perfecte-markt-scenario, althans voor dié toepassing. De implicatie voor bedrijfswaardering is dubbel. Ten eerste, indien bij een practische uitwerking DD een andere waardering oplevert dan DCF, dan moet dit te wijten zijn aan inconsistenties tussen de veronderstellingen van de twee berekeningen. Zo gaat men bijvoorbeeld bij DD vaak uit van simpele dividendvoorspellingen met, vanaf een zekere horizon, een constante groeivoet in de uitbetalingen; en men staat niet stil bij de vraag of de veronderstelde uitbetalingen wel consistent zijn met de bij DCF aangenomen vrije kasstromen en 'continuing value'. Een tweede implicatie van de veronderstelde irrelevantie van het uitbetalingsmoment is als volgt: aangezien een correct en consistent toegepaste DD en DFC per definitie hetzelfde resultaat opleveren, loont het eigenlijk niet de moeite om uit een kasstroomprojectie ook nog een overeenkomstige haalbare dividendpolitiek te distilleren en die dan te waarderen. Doe dus gewoon DCF, en brei er geen DD-analyse aan: het vraagt minder werk, en men maakt minder kansen op fouten. Deze redenering gaat op voor zowel groeibedrijven als gevestigde waarden. Terugkerend naar de vroegere probleemstelling: de meest belangrijke bron van onzekerheid over de bedrijfswaarde is de onzekerheid over de kasstromen eerder dan het moment van uitbetaling. Bij jonge groeibedrijven is de onzekerheid over de kasstromen intrinsiek groot, en het overschakelen van DD naar DCF lost dit probleem niet op. IV. DCF EN BESLISSINGSBOMEN In heel wat handboeken en artikels komt er enige onduidelijkheid voor omtrent het onderscheid tussen DCF en optiewaarderingsmethodes. Als men in de literatuur stelt dat DCF de flexibiliteit onderschat, dan bedoelt men eigenlijk vooral dat DCF meestal verkeerd toegepast wordt. DCF hoort te vertrekken uit verwachte kasstromen voor elk jaar, d.w.z. het gewogen gemiddelde over alle mogelijke kasstroomscenario's voor dat jaar - dus niet uit de meest waarschijnlijke kasstroom, en
evenmin uit het gemiddelde nadat men enkele belangrijke scenario's weggelaten heeft. In principe zou DCF ook moeten gebruik maken van beslissingsbomen. Laatstgenoemden laten ons toe te weten welke beslissing rationeel zal genomen worden in welke omstandigheden. Als dusdanig staan ze los van optietheorie. In een klassiek handboek als Brealey en Myers (1996), bijvoorbeeld, worden beslissingsbomen al jaren aanbevolen (zie hoofdstuk 10). Een cijfervoorbeeld omtrent een mogelijke uitbreidingsinvestering kan het voorgaande illustreren. Figuur 1 beschouwt een groeibedrijf dat nu (tijdstip t=O) moet gewaardeerd worden. Tijdens het komende jaar zal het duidelijk worden of het door de onderneming geproduceerde product al dan niet zal aanslaan. Het bedrijf heeft de mogelijkheid volgend jaar (tijdstip t = l ) via een bijkomende investering van 200 de productiecapaciteit uit te breiden. Deze inspanning kan, afhankelijk van het feit of het product een succes wordt, vanaf het jaar erna (tijdstip t=2) de kasstroom gunstig beïnvloeden. Vanaf dat ogenblik komt het bedrijf in een 'steady-state' toestand terecht waarbij de verwachte kasstromen constant blijven doorheen de tijd. Figuur 1 schetst het probleem, Figuur 2 toont twee waarderingsmethodes: eerst rekening houdend met de flexibiliteit, en dan zonder. Toegepast op DCF impliceert het voorgaande voorbeeld dus het volgende: eerder dan voetstoots aan te nemen dat de in het bedrijfsplan beschreven uitbreidingsinvestering er hoe dan ook komt (wellicht het meest waarschijnlijke scenario), wordt via de beslissingsboom in rekening gebracht dat die investering alleen uitgevoerd zal worden als de zaken voldoende goed draaien. En aangezien men ervan uitgaat dat de model-opsteller de kans kent dat de zaken voldoende goed zullen draaien, kan men meteen ook de correcte verwachte kasstroom berekenen die als basis hoort te dienen voor DCF.
FIGUUR 1 Data voor het waarderingsprobleem
uitbreiden: investeer 2// op t=l
I
kasstroom uit activiteit = 50,
doe niets
van het
Q vraag??
uitbreiden: op t=l
perpetueel
\ niets',
piiiGG-1 uit activiteit
tijd(, jaren) d: waarderingstijdstip
Diskontovoet = 10%
t=l: vraag wordt geobserveerd; eventuele investering
t=2, ... eventuele investering is afgewerkt verdere exploitatie
FIGUUR 2 Waurde~ingvolgens mee DCF-methoden
Oplossing met beslissingsboom 1.1s het opti~nanluit te breiden op t=l ? bij hoge vraag: netto actuele waarde kasstromen op t=l indien uitbreiding:] 320 netto actuele waarde kasstromen op t=l indien geen uitbreiding? 220 bij hoge vraag is het optimaal uit te breiden. Op ogenblik t=l is de waarde van de onderneming dus 320.
-
*
bij lage vraag: netto actuele waarde kasstromen op t=l indien ~ i t b r e i d i n ~ : ~ - 90 netto actuele waarde kasstromen op t=l indien geen uitbreiding:4 110 bij lage vraag is het optimaal niet uit te breiden. Op ogenblik t=l is de waarde van de onderneming dus 110
*
2 . Wat is op tijdstip t=O de neno actuele waarde van de onderneming, inclurief uitbreidingsmogelijkheid? 320 x 0.5 + 110 x 0 . 5 = 195 1.1
3 . War is op tijdstip t=O de netto actuele waarde van de onderneming zonder uitbreidingsmogelijkheid? 0.5x(g+&+ ...) + 0 5 x ( f i + 10 ....) = l 5 0
m+
Foute oplossing zonder beslissingsboom Doorgaans veronderstellenDCF-oplossingen die geen rekening houden met de latere keuzemogelijkheiddat de uitbreidingsinvesteringaltijd zal doorgaan, ongeacht de vraag op tijdstip t=l. Dan is de verwachte waarde van de kasstroom: vanaf tijdstip t=2 gelijk aan -200 + 0.5 x 20 + 0.5 x 10= -185. voor tijdstip t=l: 0.5 x 50 + 0.5 x 10 = 30.
voor de latere jaren:
De totale waarde van de onderneming op t=O wordt uiteindelijk berekend als: 30 1.1 + 1.12 + 1.13 + .... = 105
E
Toelichtineen: 1
(s+ s + ...)
20-ZOO+
20 + . . . 2 2 0 + ( -1.10 - + - 1.102
10 4 1 0 +1.10 ( ~ +1.102 - +
= 20-200
) = z o + -0.10 =z20
) = l o + "0.10 =llo
...
+a 0.10 = 320
V. DCF, BESLISSINGSBOMEN EN OPTIETHEORIE
Beslissingsbomen lossen echter niet noodzakelijk alle problemen op met betrekking tot DCF. In hoofdstuk 10 van Brealey en Myers (1996) wordt inderdaad meteen ook opgemerkt dat het risico, en daarmee ook de vereiste disconteringsvoet, van de toekomstige kasstromen mogelijk~afhangt van de vraag of de zaken binnen n jaar goed zullen draaien of niet. Om dit te illustreren in ons voorbeeld: het is niet noodzakelijk juist dat het risico van de uitbreidingsinvestering hetzelfde is als dit van de basisinvestering en dat dit risico doorheen de tijd op dezelfde wijze evolueert. Binnen een DCF-benadering kan men met dergelijke problemen moeilijk overweg: de standaard-veronderstelling van DCF is dat het niveau van toekomstige risico's op voorhand gekend is (en meestal zelfs constant doorheen de tijd). Optieprijstheorie lost dit probleem wél op: een optieprijsmodel houdt er rekening mee dat, als een optie 'deep out-of-the-money' geraakt, de hefboomfactor stijgt, en vice versa. Met deze variërende hefboomfactor verandert ook het risico, en houden optieprijzen impliciet rekening met door de tijd heen veranderende risico's en discontovoeten. De werkelijke bijdrage van optieprijstheorie is dus niet zozeer het gebruikvan beslissingsbomen,maar wel het berekenen van huidige waarden die rekening houden met onzekerheid over toekomstige risico's. Optiemodellen houden, in tegenstelling tot DCF, dus rekening met fluctuaties in risico. Toegepast op het voorbeeld zou men op waarderingstijdstip t=O de uitbreidingsmogelijkheid kunnen beschouwen als een calloptie met als uitoefeningsprijs het investeringsbedrag (= 200); met als 'onderliggende' de marktwaarde van de extra cashflows die men krijgt als de vraag hoog is en men de investering uitvoert; en met als vervaldatum het tijdstip t = l. Schommelingen in deze vraag tesamen met de hefboomfactor bepalen het risico van deze optie. Uiteindelijk reflecteert de som van deze callwaarde en de waarde van de onderneming zonder uitbreidingsmogelijkheden de theoretisch correcte waardering van het bedrijf dat op tijdstip t = 1kan kiezen al dan niet bijkomend te investeren. Terloops weze opgemerkt dat het risico van een optie, omwille van de hefboomfactor, verschilt van het risico van het onderliggend actief. Ingeval van een cal1 (een aankooprecht) is de positie in het onderliggend actief f'long" (men wint bij koersstijgingen) en is er een hefboomfactor (de procentuele koersschommelingen van de optie zijn groter dan deze van het onderliggend actief), hetgeen het risico ver-
hoogt en de vereiste disconteringsvoet optrekt. Een projekt met een call-achtige optie, zoals in het hogere voorbeeld, leidt dus tot een lagere waarde dan een DCF-waardering die (correct) vertrekt van een op beslissingsbomen gebaseerde verwachte kassstroom maar (incorrect) verdisconteert aan een rentevoet die geen rekening houdt met de flexibiliteit. De stelling dat optiemethodes een hogere waarde opleveren, betekent dus dat het onderschatten van de verwachte kasstroom via DCF zonder gebruikmaking van beslissingsbomen doorgaans belangrijker is dan het onderschatten van de disconteringsvoet. Of, beter nog, de verhoging van de waarde komt vooral van het gebruik van beslissingsbomen; het gebruik van optietheorie kan die geschatte waarde dan ietwat naar beneden bijstellen (bij call-opties), of ietwat verhogen (bij put-opties, zoals het recht om de fabriek te liquideren.) De door optietheorie aangereikte oplossing is echter alleen gedeeltelijk, omdat de waarde van het bedrijf-met-keuzemogelijkheden nog steeds afhangt van de marktwaarden van elk der alternatieven, inflexibel toegepast, en van het gedrag van die marktwaarden doorheen de tijd. Met andere woorden, zelfs om een bedrijf te waarderen dat mogelijks eenmalig ooit een uitbreidingsinvestering zal doorvoeren zoals in bovenstaand voorbeeld, moet men eerst reeds twee 'onderliggende' hypothetische bedrijven kunnen waarderen: ten eerste een theoretisch bedrijf dat klein begint en ten eeuwigen dage klein zal blijven, en ten tweede een even hypothetisch bedrijf dat meteen op grotere schaal begint en ook altijd met dezelfde grotere capaciteit blijft verder leven. Men moet ook weten hoe en in welke mate die twee waarden kunnen fluctueren doorheen de tijd, en hoe hun waardeveranderingen samengaan. Men moet dus nog steeds het risico van de 'onderliggende(n)' (met name de evolutie van de marktwaarde van toekomstige kastromen naargelang de vraag in het geval van hogerstaand voorbeeld) zelf definiëren; en optietheorie vertelt dan hoe daaruit het risico van de afgeleide producten kan worden gedistilleerd met name de waarde van de keuzemogelijkheid). Indien meerdere malen uitbreidingsinvesteringen mogelijk zijn of meerdere investeringsniveaus zal al gauw de complexiteit van het model en het aantal te schatten 'onderliggenden' exponentieel toenemen.
VI. FINANCIELE OPTIES VERSUS REELE OPTIES Voor het waarderen van opties op financiële instrumenten kan de analist grotendeels terugvallen op het denkwerkvan de kapitaalmarkt zelf. Financiële markten hebben vaak ook genoteerde opties, die niet alleen (en uiteraard) consensusinformatie geven over de waarde van de optie zelf, maar ook toelaten de impliciete onzekerheid te schatten en zelfs, als er voldoende verschillende opties zijn, zowat de gehele impliciete verdelingsfunctie af te leiden. Bovendien geeft de markt ook de prijzen van de onderliggende(n), vaak met lange historieken die als steun en toetssteen kunnen dienen bij het modelleren van mogelijke toekomstige koersevoluties en hun samenhang. Wissel- en goederenmarkten hebben bovendien ook termijnmarkten op diverse looptijden, die ons toelaten de vereiste risicocorrectie van deze 'onderliggenden' te schatten op elk van die looptijden, desnoods met intra- of extrapolatie. En tenslotte zijn financiële opties (weer in tegenstelling tot reële opties) doorgaans ook kortlopend. Dit is belangrijk omdat de impact op de prijsbepaling van niet-lukraak gedrag in aandelenprijzen of wissellcoersen, of meer algemeen afwijkingen ten aanzien van de bij de prijsbepaling van opties gemaakte veronderstellingen, vooral voor verre vervaldagen merkbaar worden. Voor het waarderen van 'reële' opties is de analist in een veel minder comfortabele positie: alle parameters moeten door de analist zelf aangebracht worden. Om terug te keren naar het zonet aangehaalde voorbeeld van de mogelijke toekomstige uitbreidingsinvestering: toepassen van de techniek van reële opties vereist dus het bepalen van de waarde vandaag van de twee hypothetische 'onderliggende' inflexibele bedrijven; bovendien is het ook nodig voor deze twee ondernemingen de waardeveranderingen over de tijd, en in het bijzonder de onderliggende oorzaak ervan (hier het niveau de vraag), te modelleren. Dit is niet eenvoudig, maar vaak heel belangrijkvoor de te schatten bedrijfswaarde. Zo is 'optie'-waarde van een oliebron of een kopermijn - een relatief simpel bedrijf - erg gevoelig aan de manier waarop de olie- of koperprijs in model wordt gebracht. Iedereen gaat vermoedelijk akkoord dat dit soort prijzen op lange termijn geen lukrake paden zijn, maar hoe men ze wél moet modelleren is niet evident. En, zoals gezegd, bij een probleem met een lange horizon, zoals de waardering van een groeibedrijf, is een model dat passend niet-volledig-lukraak-gedrag vat cruciaal. Bovendien is een variabele zoals stijging of daling van de verkopen die de opportuniteit van het al dan niet
doorvoeren van de uitbreidingsinvestering bepaalt, ook niet op de beurs genoteerd; er is weinig of geen empirisch materiaal beschilibaar dat als vertrekpunt kan dienen voor de modellering; en de analist heeft er het raden naar hoe het werkelijke evolutieproces van verkopen, marktgrootte, of technologische vernieuwing, moet worden gecorrigeerd voor risico-allemaal vragen die men moet beantwoorden vóór men de waarde kan berekenen van bedrijven met of zonder flexibiliteit. Niet alleen is bij een financiële optie veel meer informatie bescliikbaar in verband met de onderliggende processen, maar ook is het verband tussen de optiewaarde en de onderliggende variabele(n) goed gedefinieerd. Bij een gewone call, bijvoorbeeld, is de kasstroom gelijk aan het (positieve) verschil tussen uitoefenprijs en marktwaarde van het onderliggende. Dit is duidelijk en ondubbelzinnig. Het is veel minder evident wat de kasstroom is van de 'grote' of 'kleine' versie van een groeibedrijf bij elk niveau van de onderliggende processen als, bijvoorbeeld, marktgrootte of produktiekost. Uiteindelijk is er nog het probleem van de 'continuing value'. Bedrijven of analisten zijn om begrijpelijke redenen niet geneigd verfijnde voorspellingen te doen over de verre toekomst. Voor alles dat verder ligt dan tien jaar (of zelfs vijf jaar) wagen ze zich niet meer aan specifieke, op maat geinaakte predicties en vallen ze liever terug op een eerder mechanische 'steady-state'-voorspelling. Het probleem is natuurlijk dat, voor groeibedrijven, de bijdrage van nabije kasstromen tot de totale huidige waarde relatief klein is tegenover het aandeel van verre kasstromen. Spijtig genoeg kunnen optiemethoden evenmin als de meer traditionele dit probleem oplossen. Immers bij een optie-achtig model wordt de 'continuingvalue'geïmpliceerd door de veronderstellingen over onderliggend(e) proces(sen), de gevolgen van die variabele(n) voor de kasstromen, en de doorgevoerde correctie voor risico. Een voorzichtig analist hoort een optiewaarderingsmodel dan ook te toetsen, door bijvoorbeeld de redelijkheid van de geïmpliceerde verwachte bedrijfswaarde binnen vijf of tien jaar te beoordelen. En dat brengt ons terug tot het initieel probleem: eigenlijk is alles dat meer dan vijf of tien jaar van ons af ligt koffiedik kijken, zodat een dergelijke toetsing quasi onmogelijk is. Bovendien staat, bij een groeibedrijf, zoals gezegd de 'verre' toekomst voor een groot deel der waarde. Maar de 'verre' kasstromen zijn ook het meest onderhevig aan modelleringsfouten in elk van de vereiste stappen - modelleren van onderliggende proces(sen), het verband tussen onderliggen-
de proces(sen) en de kasstromen, en correctie voor risico. Terwijl de geïmpliceerde 'continuing i~alue'vaneen optieprijs vermoedelijk meer beredeneerd is dan het gebruikelijke natte-vinger-werk bij DCF, is het risico van (Iedeiierings)fouten wegens de toegenomen complexiteit overeenkomstig groter. Tenslotte, ter vergelijking, indien wij de onderneming van ons voorbeeld zouden waarderen met behulp van de optiemethodologie (dus modelleren via een beslissingsbooni maar daarna verder verfijnen om liet risico beter in kaart te brengen) en hierbij gelijkaardige veronderstellingen zouden aannemen als in Brealey en Myers (1996), hoofdstuk 21 blz 593-595, wordt, bij een risicovrije rente van 5%, de waarde van onze onderneming 194 in plaats van de hoger enkel op basis van de beslissingsboom bekomen 195. Dit verschil is te wijten aan het voorheen vermelde feit dat bij calt-optie-achtige flexibiliteit de optieinethodologie impliciet hogere discontovoeten toepast. Merk ook op dat het verschil tussen het resultaat van DCF-met-beslissingsbomen en optie-waardering, 195versus 194, miniem is in vergelijking met verschil tussen DCF-met-beslissingboom en (onjuiste) DCF-zonderflexibiliteit, 195 versus 105"
VII. BESLUIT De verdiensten van beslissiiigsbomen, al dan niet gekombineerd met optiemethodes, zijn onmiskenbaar. Zoals DCF in vergelijking met DD, dwingt een optiemodel de analist tot beredeneerde voorspellingen in plaats van oppervlakkig natte-vinger-werk. Bovendien vereist een optiemodel een grondig nadenken over risico en risicoveranderingen over de tijd. Toch inogen die troeven de analist of bedrijfsleider niet blind maken voor de mogelijke fouten die geïntroduceerd worden door verkeerde veronderstellingen over de onderliggende processen, de implicatie van de onderliggende variabelen voor de kasstromen onder diverse investeringsbeslissingen, en de correctie voor risico. Er is een reëel gevaar dat, omwille van de techniciteit van de modellering, de uiteindelijke gebruiker weinig inzicht heeft in het hele waarderingsproces en in de modelleringsfouten die kunnen optreden. Optieinethodes, alhoewel theoretisch superieur, leveren voor de praktijk geen wonderoplossing.
NOTEN 1. Weliswaar telden vroegere introducties in Belgic relatief weinig liigh-tech bedrijveii, maar men kan zich toch steeds de vraag stellen of liet na 1995 niet eerder gaat oni een klei11 staal van bedrijven uit een zeer korte en uitzonderlijke periode, dat sleuii heeft gekregen van eeii onverwacht gunstig beurs- en coiij~inciuurklimaat. 2. Zie Figuur 2. 3. Zie Figuur 2. 4. Zic Figuur 2. S. Zie Figuur 2. 6. in de zesde editie van Brealey en Myers (2000) kali met de bespreking vali het gebruik van beslissingsbomen opilieuw terligvinden in hoofdstuk 10 cn een behandeling vaii rccIe optics eveneens opliieuw in hoofdstuk 21.
REFERENTIES Brealey R. en S. Myers, 1996, Principles of Corporate Finance, Se editie, (McGraw Hill). Brealey R. en S. Myers, 2000, Principles of Corporate Finance, 6e editie, (McGraw Hill). Imam M. ei1 C. Van Hulle, 1998,The Post-Issue Price Performaiice of Initial Public Offerings in Belgium, Cnhiers Econonziq~iesrle Brwelles, 3e trim, 159, 279-297. Ritler J., 1984, The Hot Issue Market of 1980, Jourrinl ofBusiness, 57, 215-240. Vandemaele S., 1997,A ~ ~ c t i oaild n Auction-Like IPO Underpricing Theories, TijcEsch+t oor Economie en Mu~zagemelztXLII, 3, 311-336. Van I-Iulle C., M. Casselman and M. Imam, 1993, Initial Public Offerings in Belgium - Theory and Evidence, Tijdschl.ifl voor Eco~zor~zie en Malzagement XXXVIII, 4, 385-423;
HOTEL BINNENHOF &
BEGIJNHOF SHOTEL
Danken U voor het vertrouwen dat U ons schenkt.
e
BEKAERT STAALDRAAD
1 T E L E N S ALS JE LEVEN ERVAN AFHANGT Van alle Rekaert produktcn zijn de ha~i~lelspro~lul<~eii, eii ineer spccifiek de atrastcringen, het iiieest z~clitbaarei1 dus liet heat gekend bij een breed publiek. Bekacrt afraatcriiigen
zijii
rniioi e11 cliiiirza~in,inaar in de
allereerste piaais beschermen en hcveiligcii ;c tic iiienseri e11 hiin directe oinjieviiig. Bekacrt ataaidraa~lis niet steeds zo op\~allciidaaiiwezig zoals Het is meestal verborgcii, maar rocli vaii vitadl belaiig, in af~~steringeii. zoals staalkoord voor tie veraterkiiig \,ai1 r,idianlhantIeri, staali~ezelfiltea iii
airbags, malieiiiiig van IiijpleiilinSeii in zee, gebruik i11 coiiipiitcrs of in
kritiaclie fiitniticProcessennnn Telkcns vindt inen daariii een staaltje terug vali Relmerta leiijerschap in k ~ \ ~ ~ ~ l icn t c i ttcclinologie. De 16.000 inciicwcrkcrs va11 de Hekaertgroep iiit rneer ilan 50 ~irodukticccii~rai11 19 laiiileii eii ~iitilc veie ~crkoo~ltantoreii, gevcii iii de werel van siaalilrn,ii? ile tiinii Jali en ataaii horg voor k\wlileit ei1 I-ietrou\i~baarhci~I.
O
W E
K N O W
H O W
N.V Bekaert S A., Groepsdirectie. President Kennedypark 18, B-8i00 Kortrijk
our tradition of academic excellence 30 years of MBA experience, within a 100-yeai business school and u 575-year university 45 faculty from over 25 top European and US universities a two-seinester MBA program that still einphasizes economic anaiysis and rigor - a two-semester pre-MBA program for students without prior business education
our international focus a rigorously seiected class of l25 students from over 30 countries, and a network of 2500 Alumiii active in 70 countries exchange programs with Chapel Hili, Chicago, Columbio, Corneli, Duke, ESADE, Noriiiweslern, Peiin State. Rochester, and UC lrvine. e
a superb setting a new MBA campus thai uniqcieiy biends gothic, baroque, and moderii architecture u genuine university town fiiii of medieval splendor, gastronoiny. and life right in Europe's heart and just 30 minutes from the EU headquarters
More excuses? hitp.l/www.econ.kuleuven.ac.be/mba Gonda Huybens Naamsestraal 69, 3000 Leuven l gonda huybens@econkuleuven ac be
Tijdschrift voor Economie en Management Vol. XLIV, 3, 1999
Banking On the Eve of the 21STCentury By H. VERWILST"
I. HWTRODUCTIOW It will certainly not have escaped the impartial obseiver that the financial sector has been undergoing sweeping changes recently. At the international level, the merger was announced in April this year between Travelers and Citicorp, which will henceforth continue life as the Citigroup and be active in more or less every segment of the financial sector. To give you an idea of the scale of this new group: the market capitalisation is estimated at some BEF 5750 billion, wel1 over half Belgium's gross domestic product and 80% of the total market capitalisation on the Brussels Stock Exchange. In Switzerland, this year, the first and third largest banks in the country merged (UB§ and SBC). The new bank has a market capitalisation of 2320 billion francs and is the largest global assts manager in the world. And this week, Deutsche Bank took over Bankers Trust and Crédit Eyonnais Bank Belgium creating a group with a balance sheet of more than BEF 28000 bn or three times the GDP of Belgium. Moreover, in our own country, there have of course been the take-over of BBL by ING, the merger of KB with Cera and - last but not least - the teaming up of Generale Bank with the banking and insurance group Fortis. Operations of this kind look spectacular on account of their scale and consequently receive a good deal of media attention, but they are nevertheless merely the surface reirection of a much more deep-reaching trend or revolution, if I might use that word, in banking. Changes in the financial sector are nothing new - far from it. What is new is the
-'University of Gherit and Fortis Manageinent Board. This speech was delivered as the inaugural lexture of tlie Generale Bank Chair 1998-1999.
pace and scope of the changes we are seeing now. Al1 aspects of banking are currently subject to fundamental rethinking. II. WHY IS BANKING CHANGING? I should like to deal firstly with the causes of these changes in the financial sector, concentrating mainly on explaining the crucial role of technological development, and secondly with what the implications and challenges of al1 this are, on the eve of the 2lStcentury, for banks in genera1 and for Generale BanWFortis in particular. What are the reasons behind these major changes? Basically, it al1 comes down to the fact that a number of factors are eroding the added value which banks create when performing their key functions in economic life. This leads to an impending loss of profitability. In principle, the banks can try to cushion this in three ways: by cutting their costs, by seeking new activities and income or possibly by taking more risk, which should normally be counterbalanced by a higher anticipated yield. To illustrate this impending danger to profitability and at the same time to show that even in the banking sector profitability is not self-evident, I should like to quote tlie following figures: of the 93 banks governed by Belgian law in this country, about ten achieved a good return on own equity (over 15%) in 1997. On the other hand, in a relatively favourable interest rate climate for banks, the return on equity in 1997 of as many as 40 of the 93 banks was lower than the average yield of 5.3% on risk-free long-term government bonds. Of these forty banks, five suffered losses last year. Before focusing on the factors which gnaw away at the added value which the banking sector contributes to the economy, it is possibly a good idea just to look briefly at the question of why banks exist. Traditionally, banking fulfils three key functions in an economy. The first has to do with taking care of payments. This is important because an efficient payments system makes a not inconsiderable contribution to the economic growth of a country. The second function is the socalled transformation conversion function, whereby the bank helps its customer to tide over in time and space and assumes a number of the customer's risks. The classic example here is of course the role of intermediary played by a bank between, on the one hand a family which wishes to set aside money for a short term, and on the other hand, a business which wishes to finance an investment project in the longer term. In the more recent thinking about the role played by banking in
the economy, a very great deal of attention is paid to the third role which the banks play in ecoiioinic life and which has to do with solving the so-called problem of asymrnetric inforination. In short, this boils down to the following. A company which needs money usually knows perfectly wel1 on what and how it will spend this money. The potential lender, alias the investor, however, is far less aware of this and inevitably remains somewhat in the dark about the true intentions of the potential borrower. It would be too time-consuming and expensive a task for the potential investor time and again to assess each project on its merits and to monitor its implementation. The bank can intervene here. In this respect, is has a comparative advantage, not only because it is specialised in activities of this kind, but above al1 because it has more information at its disposal. The bank usually has a long-standing relationship with tlie undertaking, in which there is regular personal contact with the management. In this way, the bank obtains a more or less complete view of the company's financial flows and its generally better able to assess the credit risk. In a nutshell, we could say that banks exist because the world is not perfect. Banks are institutions which make it possible for economic operators to circumvent these imperfections. However, the added value which a bank can offer when fulfilling each of these functions is coming under increasing pressure, since a number of these imperfections are disappearing - for instance, the existence of eleven different currencies in eleven different European countries - and because a number of rivals are springing up who will themselves be offering services which until recently had been the province of banks. Here we only have to think of supermarkets offering loans and computer giants offering information processing. The upshot is increased competition among financial institutions, resulting in tighter margins. For this reason, banks are constantly looking for new products and new techniques to meet custoiner requirements. You only have to think in recent years of variable mortgage rates, the introduction of click funds, the expansion of electronic payments, to name but a few. There are about four major reasons for the increase in the competitive pressure and thus pressure on the banks to constantly adapt. Firstly, technological developments have thrown up a number of possibilities with respect to information processing, product distribution and risk management, which were not around until recently. The
threshold at which it is possible for non-financial institutions to develop their own financial services is thus much lower. For instance, think of large motor companies which not only have their own extensive financial seivice fulfilling the role of bank for the different units in the multinational company, but can also offer their buyers madeto-measure car financing. Technology has also meant that the optimum scale for a number of activities has increased. One example of this is the administrative processing of financial transactions, known as back-office operations. In order to make the best possible use of the technology available, a high enough volume of transactions must be processed so that the cost per transaction is reduced. In the present circumstances, those who do not reach the required volume often suffer a cost disadvantage compared with larger competitors. I shall return to the role of technology in the future development of banking in a minute. Allow me first to mention a number of other causes of the change in the banking landscape. Secondly, in addition to technology, the different perception of the role which governments should play in the financial markets has had a considerable impact. Governments traditionally had a larger finger in the financial marltets pie, whether directly via public credit institutions which dominated a number of market segments or more indirectly by regulating both the financial institutions and the working of the markets. This strong government interference was prompted by three considerations. Firstly, in conducting their business the banks rely on a substantial leverage effect because they mainly work with borrowed funds, i.e. the deposits entrusted with them, which are matched by relatively low own funds. Since most depositors are not in a position to check whether their bank is not taking too much risk with their money, the government took this task upon itself. The second reason is closely connected to this. Banks are specialised, among other things, in trade on differing time scales, i.e. as I said just now, they bridge over time, as it were, by converting short-term deposits for example int0 long-term loans. Confidence is consequently one of the crucial elements for the soundness of an existing financial order. If one bank gets int0 difficulties, it is highly likely that people wil1 not only lose confidence in the one bank, but also in the other banks. This may then lead to a genera1 run on the banks, resulting in the entire financial system and also the real economy ending up in crisis. We do not need to look far nowadays for examples of economies where no confidence in the currency and insufficient solvency of the financial
institutions has led to enormous economic problems. To try to avoid this system risk (which has not always been successful in the past, but that is another story), the governnlents of most countries have intervened with strict regulations. Finally, it is also true that the financial sector was considered for a long time as an instrument which could prove useful to conduct a social and industrial policy. The government tried through intervention to direct lendilig int0 the projects it favoured. It is hardly surprising that in such a climate of regulation and government intervention, the corporate culture of commercial banks mostly gave priority to stability and status rather than innovation, creativity, speed and entrepreneurship. In most cases there was no clear costbenefit analysis per product or customer so there was considerable cross-subsidisation. Surpluses in one segment were used to offset losses in others, without anyone having a clear idea of the extent of such measuses. As far as international trade in goods and the functioning of the commodities markets are concerned, considerable efforts have already been made to open them up, thereby promoting economic growth. However, in order to reap al1 the benefits of this, it was necessary to open up financial services as well, both on the home market and at international level. The latter occurred within Europe through the Second Banking Directive, which, provided a number of conditions were met, allowed a bank in another Member State of offer its services under home country control. In addition, but by no means of secondary importance, there was also a more practica1 reason for government withdrawal from the financial sector. The psobleins faced by public finances meant that in most countries the government was wel1 able to use the proceeds of a possible privatisation of the public credit institutions. Moreover, the situation was such that the government did not have the resources to bolster the capita1 basis of these Stateowned institutions in order to allow them to keep pace with the expansion of their activities. A so-called "modernisation" of the domestic money and capita1 markets als0 had to cut the cost of financing the government debt by making the markets more competitive and lowering the profit margins of the financial institutions. The combination of the opening up of the financial markets and the new possibilities offered by technology to "unbundle" products or services or to split them int0 components was a direct attack on cross-subsidisation. New competitors within and outside the finan-
cial sector even started cherry picking - that is, they picked the cherries out of the pie of the established institutions by targeting the most profitable activities and left the rest - namely the in some cases lossmaking activities, such as payments - to the others. Thirdly, along with technological advances and the opening up of financial markets, demographic trends have also worked somewhat against the comparative advantages of banks. The ageing population, together with higher average levels of education and a standard concentration of wealth in the higher age ranges, has meant a shift in saving behaviour, whereby investment consultancy services are gaining in importance to the detriment of taking savings deposits and granting credit facilities to individuals. The result is that a lower portion of financial flows is booked through the banl<s7balance sheets. This trend is being further reinforced by the increasing irnportance of capitalisation systems for pension savings, which gradually (must) come to supplement the distribution system. This has meant the advent of new institutional investors attracting capital. The banks admittedly came up with a solution here by themselves taking initiatives in the institutional savings sector, but the remuneration here is quite a bit lower that for traditional banking activities and it is not certain whether the banks wil1 in fact keep a comparative advantage here. Fourthly, we must of course consider the introduction of the euro. Whereas what I have said so far applies in varying degrees to banks throughout the world, the advent of the single European currency presents an additional challenge to European banks in particular. The EMU project is already driving the process of change in the financial sector, and this will become more marked in the future. For one thing, the introduction of the euro means that some of the income from foreign exchange transaction will disappear. More importantly, the protective barriers behind which banks could up to now shelter on the national rnarket will disappear in one fel1 swoop. With the introduction of the single currency, Belgian banks will no longer have a natural currency advantage for intermediation in financial flows between residents. The home market for the Belgian banks will gradually become the entire euro area, at the start possibly primarily for the professional rnarkets and only to a lesser extent for the retail segment. For that matter, the companies of other economic sectors will also take a more pan-European approach in the future and will thus expect their bank to be willing and able to follow suit.
Separately and together these four factors - techiiological developments, the opening up of the financial marltets, demographic trends and the single currency - will radically change the face of banking in the years to come. 111. THE ROLE O F TECWNOLBGY: A CLOSER LOOK
Tlie crucial question is of course where these developments will lead banking and what is at this moment the most recommended course or courses - of action. As I promised just now, I am coming back to the role of technology, since it is one of the most important factors determining the developments in the near future. It would be unreasoriable to attempt to give a precise forecast of whal technological developments will occur within the next five years and to try to define a precise timetable. Note that it is not impossible that, on account of the screening of IT systems and the adaptations which will nevertheless have to be made to certain software applications in the banking sector as a consequence of the introduction of the euro, the year 2000 problem and the implementation of mergers, new technologies could be incorporated int0 the banks' IT packages somewhat faster than would othenvise have been the case. The Belgian banks' IT investments in recent years have been enormous. Of the some 100 billion francs which the banks have spent on investrnents in the past four years, an average of some 40% to 45% was earinarked for information technology. The technological advances in respect of payments has also been very rapid for the ordinary man in the street. From the introduction of automatic cash dispensers at the end of the 1970s, from which customers can use their bank card and a secret code to draw cash and check the balance in their account, to payment terminals at petrol stations and later in department stores and other businesses, to the selfbank, where the customer can deal with a wide range of banking transactions himself. In addition, in recent years, new distribution channels have been on the up and up. Telephone banking and banking via PC have rapidly become established, especially among younger people, but also at companies via Isabel, the interbank system for electronic payments. Moreover customers are being gently led, as it were, to forms of payment which entail the lowest costs for the banks. In this connection, the figures for the first half of this year speak for themselves: the number of cheques, which are traditionally very expensive
for the banks to handle (the total cost of a cheque for the bank amounts to about 100 francs) was 11.3% lower than during the same period last year. The number of transfers by telephone rose by 10.9% over the same period and the number of transactions using payment cards by as much as 26.5%. This last figure naturally has a great deal to do with the abolition of charges for this service in department stores and most shops. In addition, the Proton card has secured its place in the payments sector relatively quickly. And still the developments in tlie payments sector are not at an end. "Stored value" cards, such as Proton, will be given new impetus for the spread of the intelligent telephone with which the Proton card can be loaded at home. Slightly further in the future, it will be common to use such cards on PC too, as a result of which the Proton chip can be loaded with cash at home and can be used for direct "spot" payments when buying via the Internet, for example. Although this now seems more like science fiction, it could become reality in the foreseeable future. And now we have come to the Internet. Obviously, this new medium will also play a greater role in the future, not only in payments flows, but probably also in many other areas of banking. We can cite one example - the largest UK tour operator Thomas Cook offers customers the facility of booking flights or hotels on-line, and customers can also order tourist guides for t-he country they intend visiting and foreign currency. This is another instance of how non-financial institutions are starting to move int0 the field of banking. Internet banking itself is still in its infancy in our country, but serious efforts are already being made in this field. Since November last year, Generale Bank has been offering customers the possibility of using the Internet to check the balance in their account, monitor the movements in this account, make transfers, send e-mails and consult the Reuters financial information. The new distribution channels have obviously had an impact on the activities of the branch network. It is already estimated that some 40% of transactions or contacts between customers and banks no longer occur via the branch network. This could increase to 60% in the years to come. The growth in electronic payments is of course a two-edged sword for banks. On the one hand, it eases tlie financial burden of its unprofitable payment traffic. On the other hand it is easier for a foreign bank to acquire customers on the domestic market since "bricks and mortar" investment in a local network is no longer an absolute necessity. Virtual banks are perfectly feasible for lending and offer-
ing savings products too. As regards credit decisions, techilology has led to the development of scoring models, for example. These models allow identification of the crucial parameters at the basis of the decision on whether or not to grant a credit facility. Simple forms can then be designed, which may possibly be completed on-line, which allow the creditworthiness of the potential borrower to be analysed on the basis of a nuinber of standard questions. Under certain circumstailces, this can replace a personal interview without too many probleins. The first application of this could be in consumer credit, where relatively smal1 amounts are involved, but an extension to SMEs is always possible, certainly when combined witli systems of video conferences where a certain visual contact - albeit at a distance - is still possible so that the discussion runs more smoothly. In the consumer credit field, the banlis may wel1 be subject to ever keener competition from department stores and manufacturers of consumer durables, which will continue to turn out credit formulas tailor-made to the customer in very specific niches. Nevertheless, a potential conflict may arise here for a department store. What will the attitude of a department store be, for example, if a customer shows up who wishes to buy on credit but who is found not to be creditworthy? The refusal of credit may mean that the customer is lost forever, even though his financial problem is only temporary. In addition to the fields of payments and lending, information technology will give rise to significant innovations in the investment sector. To cite but one example - information on prices and financial information is already available to anyone on Internet virtually in real time. Banks no longer have the monopoly on providing investment advice1. But it is not impossible that in the course of time software providers or others will set themselves up as information providers on the Internet, thus fulfilling a kind of "Test Aankoop" function whereby they systematically sift through the electronic offers and quotes of financial institutions and markets to find the "best buy". This does not necessarily concern simple, standardised products. Not only will investors subscribing to these services be able to save much of the expense involved in shopping around, but the information provider could also provide a link so that with a single click on the mouse investors can visit the relevant site to carry out transactions. The level op transparency which is often still the privilege of professional markets could thus come within the reach of individual investors. In such a scenario,
banks would lose a substantiai part of their advisory function and precisely that part of their customer base which is the most profitable to them. In short, al1 this means that partly as a result of technological progress, customers will increasingly shop around and wil1 be less likely to retain linlts with only one financial institution, as has often been the case to date. This implies that in many cases the long-term relationsbip between banks and their customers will come under pressure, which could sometimes prove to the customers' disadvantage in difficiilt times. Customers will also want more tailor-made products and services which meet their precise requirements. This implies that banks wil1 have to take care to choose the right distribution channel according to both the needs of the customer and the costs for the bank. Fortunately, technology also offers the facilities to achieve this. But with the conlpetitor in many cases just a click of the mouse away, this obviously means that banking will not become easier in the years to come.
IV. GEULLENGES FOR BANKS The question arises which attitude a bank is going to take in face of al1 those changes. The challenge for the banks is achieving a "superior performance". This is also, for that matter, ultimately the challenge for management as well. According to Porter, superior performance is the result of operating efficiency and strategie positioning. Strategicpositioningmeans that you look at how you can do other things than your competitors os bow you can set about doing the Same things differently. The essential question here is how you can keep your coinpetitive edge the longest, because in many cases nothing can be copied faster than an original idea. Operating eficiency means that you do the Same things as your competitor, but better or, in other words, that you achieve a greater output with the Same input or the Same output with a smaller input. This can be achieved in various ways: internally, by keeping a close watch on and control over the trend in costs andlor by selling surplus production capacity to third parties. For a bank, the latter is possible, for example, for the management of car loans on behalf of motor manufacturers or for consumer cred-
it on behalf of wholesale distributors, etc. Diversification too, where a bank opens up its distribution channels to sell insurance products, for example, may offer a solution. exterrzally, this is possible by creating networks, for example via outsourcing, or entering int0 joint ventures (Bankvan de Post), or even by cooperating with a number of banks for specific projects (Banksys, Isabel, Proton). Before I go into the question of how the link-up between Fortis and Generale Bank falls int0 the above scheme of things, I would like to consider some of the more genera1implications of al1 this for the banking sector as a whole.
I have already dealt with the search for the inost appropriate distribution channel for each product and ciistomer, but banks must make a fundamental choice - whether, for instance, they wish to only offer their customers their own-brand products, or whether they will also open their distribution channels to competitors' products. Should they advise a customer to buy a competitor's product because it is more closely suited to the customer's needs than the own-brand product? Marketing will very likely have to take a more pro-active approach. Technology has made it easier to seek out and target segments of the population actively (customer or non-customer) which have possible come to a turning point or a breaking point in their lives and consequently are possible the most amenable to changing bank or to showing interest in a specific service. Such moments include inarriage, buying a home, moving, etc. And al1 banks will have to make further efforts to retain the loyalty of their existing customers and therefore work on customer relationships. There will als0 be an enormous impact on human resources management. Firstly, in the present circumstances, and in the perspective of the changes under way which I have outlined, there are, generally speaking, too many people working in the banking sector in Belgium and, what is more, the job content of many people in the sector is changing fundamentally. In an increasingly customer-driven market, the diversity and complexity of products and services is increasing constantly. People involved in advising customers must be given adequate training to enable them to speak knowledgeably when dealing with them.
A further increase of scale in the banking institutions appears inevitable. Does this means that bigger is always better? No, but it does mean that having a certain minimum size is a necessity to be able to deal with certain business - for instance, offering services to multinationals - or to do this in a cost-effective way. The empirica1 literature on advantages of scale and advantages of diversification in banking is, however, far from unequivocal about the optimum size of a bank. The most successful banking institutions will probably be those who establish and maintain a sound balance in their corporate culture between flexibility and stability and between values which boost creativity and those which work towards helping continuity. This will not be an easy task, since every banking institution groups together different activities which each has - and must have - its own culture. Retail banking requires a totally different approach from investment banking or pure trading in a market room. International diversification also increases the need for intercultural and organisational flexibility. The increasing complexity and heterogeneity of financial institutions, together with the increasingly rapid markets, increase the need for sound, effective internal monitoring procedures. Technological advances also mean new security risks. One of the greatest nightmares for a banker is possibly hacking into the computer system, with the result of intruding into the privacy of the customers or tampering with the figures. The changing banking environment also means higher demands in respects of supervision of the banking sector. The risks have undeniably increased, and the regulations of the supervisoiy bodies must not discriminate in favour of banks to the detriment of non-banking companies involved in the Same business activities. The increasingly international nature of activities requires further coordination and harmonisation of the regulations in different countries. A great deal has happened in this respect in recent years. Clear regulations are beneficia1 to banking, on condition that they do not impose any unnecessary burdens. V. GENERALE BANK AND FORTIS The teaming up of Generale Bank and Fortis fits perfectly int0 the blueprint just outlined of operating efficiency and strategic positioning.
This link-up has resulted in the creatioii of one of the largest suppliers of integrated financial services in the Benelux countries, with a sound European base. This ensures a strong foundation for further international expansion, especially in Europe, the United States and Asia. Wit11 market capitalisation of some 1200 billion francs, the new combine is one of Europe's fifteen largest financial institutions. In terms of total assets - 12,000 billion francs - it is one of the top 20 European banks. The combine has more than 62,000 employees throughout the world. The whole will no doubt be better than each of its parts in making it possible to take up the challenges facing the financial sector, which I have already mentioned. The integration of tlie banking business as from 1January 1999 wil1 as far as possible occur on the basis of the various activities or business lines. The practica1 aspects are being worked out at present within a number of working groups. Their main task is to take stock of points for attention within their fields and proposiilg priority actions, together with a timetable. One of the most important tasks of each team consists in identi@ing the cost-benefit syne r g i e ~of the integration. Bringing this integration to a successful conclusion, which is expected to take four or five years, is an enormous and exciting challenge. The operation is expected to result in total cost and income synergies estimated at 27 billion francs by the end of 2002. As a bank we are therefore already very busy preparing for the future. We have already made major efforts in I T including developing iiitegrated computer systems and setting up a cross-border infrastructure for payments and cash inanagement. In financial risk management, we have undertaken a thorough examination of al1 aspects of lending activity and activities involving the financial markets and suggested changes. As far as customer service is concerned, the range of tailor-made products and services will be extended, the distribution channels will be extended and diversified, and staff will be offered more intensive training. At Fortis group, our starting point is that banking forms a part of an integrated financial service, together with investment and insurance business. Thus Fortis offers a complete range of insurance, banking and investment products to individuals, companies and institutional investors in the Benelux countries through a variety of distribution channels. In other words, in the Benelux countries our customers have access to Fortis products through banks, insurance bro-
kers and investment advisors, or even through new media such as Internet. In its other markets, the Fortis group concentrates on selected customer seginents, product lines andlor distribution channels. At the Same time, the ambition of the Fortis group and al1 the companies in the group is to be a solid, reliable partner offering Ilexible solutions aimed a the requirements of individual customers. Mowever big we may be, the customer is still the most important factor. h d customers can count upon the joint know-how and experience of al1 Fortis group companies.
In this address I have tried to give an idea of the enormous changes taking place in banking. However, the picture is still unclear in parts since it is difficult to assess the extent and pace of certain developments. But one thing is certain - the Fortis group is preparing to face the challenges of the next century and looks to the future with confidence. NOTES
1. Possibly they do to a certain extent for the final execution of transactions. hut for how much longer?
Tijdschrift voor Economie en Manageinent Vol. XLIV, 3,1999
BOEKBESPREKINGEN
Quality in Action Patrick k. T O W S E N D amd Joan E. GEBIHImPPT (Wiley 86- Soms, New York (1992) xxl, 262 blz.)
"Quality in Action" is een typisch Amerikaans boek, dat op een alternatieve manier poogt om een leidraad aan te reiken om tot kwaliteit te komen. De basisidee van het boek wordt gevormd door de stelling dat er geen rationele manier bestaat om tot kwaliteit te komen: iedereen heeft zijn eigen ideeën omtrent kwaliteit en iedereen heeft gelijk, maarX8.heeftook heel wat over hei hoofd gezien. Dit wordt geïllustreerd aan de hand van een oude fabel waarin drie consultants door de koning eropuit gestuurd worden oin te ontdekken hoe een olifant eruitziet omdat beweerd werd dat elk belangrijk land olifanten bezat. In een naburig land was er een olifant en de drie consultants haastten zich om zo vlug mogelijk verslag uit te brengen bij de koning. Ze kwamen 's nachts aan in de stad waar die olifant zich bevond en gingen overnachten in de herberg. In hun haast om eerst terug bij de koning te zijn, slopen ze elk in het holst van de nacht naar de olifantenstal. Op de tast onderzochten ze de olifant, maar telkens werden ze opgeschrikt vooraleer ze de olifant volledig konden onderzoeken. Terug bij de koning brachten ze verslag uit: de eerste tekende een olifant met een enorme staart (hij had toevallig de staart betast), de tweede tekende een olifant met enorm grote poten (hij was niet verder gekomen dat de poten) en de derde tekende een olifant met een buitenproportioneel lijf (hij was onder de indruk gekomen van de kracht van de olifant). D e drie tekeningen waren serieus verschillend en de koning was ontevreden want hij wist nog niet hoe een olifant er nu precies uitzag: elk van de consultants had wel een belangrijk deel van een olifant beschreven, maar drie verschillende tekeningen waren het resultaat. Per toeval had zelfs geen enkele consultant de slurf ontdekt! Zo is het ook in kwaliteit, schrijven de auteurs: ieder heeft zijn eigen stokpaardje en probeert volgens eigen inzicht de kwaliteit te verbeteren. De auteurs pogen echter een iets gebalanceerder beeld te geven van het kwaliteitsdenken en bouwen dan
ook hun boek op in drie grote delen: leiderschap, participatie en meting. Voor elk van deze steunpilaren van kwaliteit worden eenendertig axiomas gegeven en op een leuke wijze geïllustreerd. De resulterende drieënnegentig axiomas vormen volgens hen dan ook een volledig overzicht van het kwaliteitsdenken. De auteurs lopen hier natuurlijk het gevaar dat ze zich bezondigen aan dezelfde fout van de drie bovenvern~eldeconsultants: misschien hebben ze slechts drie poten van de olifant onderzocht en kan hun olifant hinken in plaats van lopen. Ik laat de lezer graag zelf beslissen of dit in dit boek het geval is. Persoonlijkvond ik tijdens het lezen dat deze opbouw van het boek in axioinas met korte bijbehorende uitleg niet optimaal was omdat er geen rode draad doorheen het boek loopt: verschillende interessante gedachten omtrent kwaliteit worden in hapklare brokken vlug ingeslikt, maar het materiaal blijft niet beklijven. Ik raad dan ook de lezer aan het boek veeleer als scheurkalender te gebruiken: lees elke dag één exioma met bijbehorende uitleg en laat die gedachte een dag lang in uw hoofd rondspoken. Na een drietal maanden zult u dan een andere visie op kwaliteit verkregen hebben. Erik DEMEULEMEESTER K.U.Leaiveaa Japanese Multinationals in the Gliobal Economy. New Horizoiras in Hntermatioiraali Business P a d W. BEAMPSW, Andres DELIOS and Donals J. LECMW (Edward Elgar, Chelteaaham, Nortlhamptoai (1997)) Dit boek werd geschrevenvanuit de vaststelling dat de kwaliteit van gegevens beschikbaar voor de studie van multillationale ondernemingen meestal niet goed is. Alvast één reden daarvoor is dat regeringen enkel gegevens verzamelen m.b.t. het eigen territorium, en dus geen plaats inruimen voor wereldwijde netwerken van bedrijven, waaruit multinationale ondernemingen nu één keer bestaan. Een aan de publicatie van dit boek voorafgaande belangrijke poging o m dat gebrek aai1 gegevens weg te werken was het zogenaamde Haward Mullznatzol~al Enteyrise Projext (HMNE). Dit project resulteerde in 1975 in een gegevensbank over de ontwikkeling van dochterondernemingcn van belangiijke multinationale ondernemingen met hoofdkwartier in de Verenigde Staten, Europa en Japan. Dit boek wil eveneens op een belangrijke wijze de beschikbaarheid van gegevens voor de studie van multinationale ondernemingen vergroten. I-Iet wil o p eenvoudige wijze aan een groter publiek een weinig bekende maar waardevolle bron van informatie over Japanse multinationale ondernemingen voorstellen. Het gaat o m Krrigai shinshutsu Icigyô sôralz (Japanse overzeese investeringen), een jaarlijkse publicatie van de belangrijke Japanse uitgeverij Tôyô Keizai. D e editie van 1994,
bijvoorbeeld, bevat informatie over meer dan 15.000 dochterondernemingen van Japanse bedrijven. D e structuur van dit boek is vrij eenvoudig. Het is gebaseerd op een willekeurige selectie van 5.843 dochterondernemingen uit de editie van 1994. Voor elke dochteronderneming bekijkt men 10 variabelen: gastland en -regio; maand ei1jaar van oprichting; industrietak; de zogenaamde mode of entiy (wholly owned subsidiaiy established as greei~fieldoperation; joint venture established as greenfield operation; acquisition; dapital participation); verkoopcijfers (1993); bedrijfskapitaal evaluatie - door d e bedrijfsleiding - van de bedrijfsresultateii tijden het jaar 1993 en, voor dochterondernemingen die niet volledig eigendom zijn van de iiiultinationale onderneming, de eigendomsstructuur. Vervolgens worden de verschilleiid e variabelen met elliaar in verband gebracht. H e t eerste hoofdstuk, bijvoorbeeld, bestaat ondermeer uit een tabel die de datum van oprichting relateert aan industrietak, en dus een overzicht biedt van veranderingen door de tijd heen in d e sectorale spreiding in Japanse investeringen. D e auteurs proberen dan patronen te ontdekken en hypotheses aan te bieden o m deze patronen te verklaren. E r wordt ook steeds vergelijkingsmateriaal aangeboden door de opname van tabellen gebaseerd o p het H M N E uit Curlian, Davidson en Suri, R. (1977). Enige voorzichtigheid is geboden bij de interpretatie van de aangeboden tabellen. Vanzelfsprekend bevat de editie van 1994 van de E y ô Keizni gegevensbank enkel gegevens over dochterondernemingen die actief waren in 1993. E r worden geen gegevens opgenomen over dochterondernemingen die weliswaar ooit opgericht maar sinds reeds gesloten werden. D e E y ô Keizai gegevensbank onderschat dus d e werkelijk plaatsgevonden investeringen. D e gegevensbank biedt ook enkel een overzicht van oprichtingen van nieuwe dochterondernemingen, niet van latere kapitaalinjecties in bestaande dochterondernemingen. Belangrijker is misschien wel dat - zoals de auteurs grif toegeven - waarden genoteerd voor het jaar 1993 voor een aantal variabelen - zoals verkoopcijfers, bedrijfskapitaal, totaal aantal en aantal buitenlandse werlineiners - net dezelfde zijn als waarden genoteerd voor deze variabelen op het ogenblik van de oprichting van de dochteronderneming. Heel concreet betekent dat dat geen conclusies kunnen getrokken worden rond, bijvoorbeeld, de relatie tussen datum van oprichting en eigendomsstructuur. Het is immers mogelijk - en werd ook reeds geopperd - dat Japanse multinationale ondernemingen misschien wel dikwijls starten met een minderheidsparticipatie maar vervolgens hun kapitaalsaandeel gevoelig opdrijven, iets wat uit d e tabellen in dit boek niet af te lezen zou zijn. Gelijkaardige problemen doen zich voor met d e relatie tussen de datum van oprichting en grootte van de dochteronderneming, gemeten in termen van verkoopcijfers, kapitaal of werkgelegenheid. E r rijzen ook vragen rond het aangeboden H M N E vergelijkingsmateriaal. Heeft het veel ziii een gegevensbank afgesloten in 1993 ( E y ô Keizai) te vergelijken met één afgesloten in 1975 (HMNE)? Vertrekt men bij het doorvoeren van dergelijke vergelijking al niet van de veronderstelling dat er verschillen bestaan tussen Amerikaanse en Japanse investeringen die onafhanlielijk zijn van tijd en ruimte? Het lijkt dank ook beter te speuren naar patronen binnen het geheel van Japanse investeringen, eerder dan Japanse met Amerikaanse investeringen te vergelijken. E n o p dat gebied biedt dit boek heel wat interessante informatie aan, waarvan we hier slechts enkele voorbeelden kunnen geven. Bijvoorbeeld, de sterke concentratie van Japanse investeringen in Azië, maar ook in een relatief klein aantal landen -Duitsland en Groot-Brittannië, Brazilië en Mexico; Liberia en Sau-
di-Arabië - in andere regio's. De sterke concentratie van Japanse investeringen in nijverheid en handel. Wet feit dat Japanse multinationale bedrijven de oprichting van nieuwe verkiezen boven de overname van bestaande bedrijven. Of dat ze uiteindelijk naar volledige eigendom van de dochteronderneming tenderen. Dat, hoe hoger de verkoopscijfers en het bedrijfskapitaal zijn, hoe hoger de eigendomsgraad is, wat de zogenaamde escalating co~nmitmentidee zou onderbouwen. Of dat in Europa de verkoopscijfers hoog liggen maar het bedrijfskapitaal laag ligt, wat het zogenaamde screwdriver karakter van Japanse investeringen in Europa zou illiistreren. Wat de lezer vooral treft in dit boek is de grote diversiteit van Japanse investeringen, hun afhankelijkheid van tijd en ruimte, en de moeilijkheid om tot één samenhangende theorie van Japanse investeringen te komen. En dat is iets waar de a~iteursduidelijk moeite mee hebben. Vaak kunnen ze geen hypothese aanbieden om een bepaald patroon te verklaren, of is de aangeboden hypothese nietszeggend. Soms is ze ook duidelijk niet van toepassing, wat de indruk welt dat de auteurs op een al te eclectische manier tewerk zijn gegaan, zonder veel inzicht in de materie. Dat blijkt alvast heel duidelijk uit de pogingen van de auteurs een verklaring te bieden voor het hoge aandeel van Brazilië in de Japanse investeringen tijdens de jaren '50 en vroege jaren '60. De auteurs veronderstellen dat Brazilië en Canada) stradit verband houdt met de zogenaamde ABC (A~~stralië, tegie van Japan om grondstoffen te bemachtigen. Volgende tabellen doen de auteurs hier dan weer op terugkomen. In hun discussie van de tabel die verkoopscijfers relateert aan het gastland noteren ze dat de verkoopscijfers eerder laag liggen in Brazilië wat hun eerdere hypothese niet onderbouwt. Een zelfde conclusie bereiken ze in hun discussie van de tabel die bedrijfskapitaal relateert aan het gastland. Er wordt echter geen alternatieve hypothese aangeboden. In werkelijkheid is l-iet zo dat de Japanse investeringen in Brazilië van de jaren '50 en vroege jaren '60 geen verband hielden met de exploitatie van grondstoffen. Het ging om het opnieuw oprichten van overzeese filialen door de zogenaamde trading firms en textielfabrieken. Het bovenstaande toont aan dat het in dit boek gaat om een voorzichtig te hanteren inleiding. Bij het interpreteren van een bepaalde tabel heeft men best oog voor de totaliteit van het boek, en de beperkingen van de gegevensbank en de auteurs. Eventuele inzichten moet men dan controleren en aanvullen met detailstudies die oog hebben voor tijd en ruimte, waarbij de uitgebreide bibliografie zeker hulp kan bieden. REFERENTIES Curhan, J.O.; Davidson, W.H. en Suri, R. (1977), Tracing tlie Multinationals: a Sourcebook on US-Based Eiiterpriscs, (Cambridge, MA, Ballinger).
Henri DELANGWE K.U.Leuven
Tijdschrift voor Economie en Management Vol. XLIV, 3, 1999 EINDVERHANDELINGEN
Ethisch beleggen in de Benelux
DE BWBANDEHPE Hillde (K.U.Leuven, Lic. TEW (1999)) De laatste jaren wordt er in de maatschappij meer en ineer aandacht geschonken en ethisch handelen. Deze tendens is ook de financiële weaan milie~~vriendelijlc reld niet ontgaan wat heeft geresulteerd in allerlei ethisch-financiële producten. In België bijvoorbeeld is het aantal ethische beleggingsfondsen bijna verdubbeld in één jaar. Maar wat houdt ethsch beleggen precies in? En duidelijke omschrijving van een ethisch spaar- of beleggingsproduct in de Benelux, kwam van Dirk Coecltelberg (Blom e.a. (1996)). Hij definieerde het als een financieel product in de spaarof beleggingssfeer dat via erkende distributiekanalen van bankprodcuten worden aangeboden en op een structurele, duurzame en openlijke manier verbonden is met een ideëel doel. Met het ideële doel bedoelt men het maatschappelijk belang dat verder reikt dan het indirect of direct bevredigen van de korte of lange termijn financiële behoefte van de belegger. De ethsiche belegger zal dus naast de klassieke risico- en rendementsmaatstaf nog een derde maatstaf hanteren. namelijk het ethisch gehalte van de projecten waarin het geld wordt geïnvesteerd. Bepaalde mensen geven immers de voorkeur aan een sociaal verantwoorde besteding van hun geld waarbij het gerealiseerde rendement toch nog in verhouding staat tot het genomen risico. Mijn verhandeling werd toegespitst op de ethische beleggingsfondsen, die een belangrijk deel van het ethisch beleggingsltapitaal in de Benelux vertegenwoordigen. Zowel Belgische als Nederlandse ethische beleggingsfondsen werden aan een risico-rendementsanalyse onderworpen. De ethische fondsen uit liet Groot Hertogdom Luxemburg werden bewust uit de analyse gehouden omdat het aanbod ervan bijna uitsluitend door filialen van buitenlandse banken bleek te gebeuren. Eind 1998 waren er zowel in België als in Nederland tien ethische beleggingsfondsen. Ondanks hetzelfde aantal ethische fondsen, situeert de Nederlandse markt zich een stuk verder in de product- of levenscyclus van ethische-financiële producten dan de Belgische markt. Dit wordt gemeten door het ethisch kapitaal per capita. Zo bedraagt het gemeten ethisch kapitaal per capita in Nederland 3980 BEF, terwijl het in België slechts 976 BEF bedraagt. De Fiscale Groenregeling in Nederland is één van de verklaringen van de grotereinteresse in ethische beleggingsmogelijkheden. Op grond van deze groenregeling zijn rente- en dividendinkomsten ontvangen van een 'groene instelling' vrijgesteld van roerende voorheffing voor de eerste 1000 NLG voor ongehuwden (2000 NLG voor gehuwden)?. Deze groene instellingen hebben minimum 70% van hun balanstotaal uitstaan bij groene projecten die een gunstig effect op het milieu hebben.
De omvang van de ethische fondsen is nog geen garantie voor hun financiële perforinantie. In de empirische studie werd de performantie gemeten van acht ethische fondsen, vier Belgische (Bacob Stimulus Dynamic, Interselex Equity Euro Job, KBC Econ Fund, VMS Luxinter Ethifond) en vier Nederlandse (Andere Beleggingsfonds, ASM Aandelenfonds, ABN AMRO Groenfonds, Meerwaarde Beleggingsfonds) fondsen. Deze fondsen bestonden op 1 jailauri 1999 minimum één jaar. De risk-profit analyse werd opgesplitst in twee delen. Enerzijds werd de performantie van de ethische fondsen en hun aangepaste benchmark individucel geanalyseerd, anderzijds werd een portefeuille van deze acht ethische fondsen gevormd en de financiële prestatie van deze portefeuille werd vergeleken met een portefeuille van de acht overeenkomstige benchmarks. De individuele benchmarks werden zo samengesteld dat ze ongeveer een zelfde asset allocation en geografische spreiding hadden dan het overeenkomstige ethische beleggingsfonds. Het Belgischc BACOBStimulus Dynamic bijvoorbeeld, krijgt een samengestelde benchmark die voor de ene helft bestaat uit de MSCI World Index en voor d e andere helft uit de JPM Global Government Index omdat dit fonds ongeveer evenveel in aandelen als in obligaties belegt. Om de performantie van de fondsen en hun benchmarks te kunnen meten werd gebruik gemaakt van het gemiddeld maandelijks rendement, de standaarddeviatie, de Sharpe, Treynor en Jensen maatstaf. Deze laatste drie één-parameter performantie-maatstaven passen het rendement aan met het risico. Uit de analyse per fonds kunnen we besluiten dat zowel de Belgische en Nederlandse ethische fondsen het iets slechter dan de markt deden wanneer het totale risico in aanmerking werd genomen. Indien enkel rekening werd gehouden met het systematisch risico, presteerden de fondsen iets beter dank de markt. Deze uitspraak moet met voorzichtigheid worden behandeld. Slechts acht fondsen werden over een vrij korte tijdsperiode geanalyseerd. Er werd ook geen rekening gehouden met de eventuele correlatie met andere niet-ethische effecten1. Een laatste punt dat hierbij de aandacht verdient, is dat er geen rekening wordt gehouden met de implicaties van belastingcn p het rendement. Het ABN AMRO Groenfonds is namelijk vrijgesteld van roerende voorheffing omwille van de fiscale groenregeling, terwijl andere fondsen niet van deze vrijstelling genieten. Het onderzoek dat uitging van de portefeuille van ethsiche beleggingsfondsen en de portefeuille van benchmarks wierp een heel ander licht op de situatie. De ethische fondsen gaven in 1998 maandelijks 0,054% extra rendement op de markt per eenheid totaal risico. Bij de gewone buy en hold strategie zou de sociaal verantwoorde belegger in 1998 0,212% meer rendement verdiend hebben dan d e markt per eenheid totaal risico. Ondanks de beperkte resultaten van deze performantie-analyse, hebben we toch bewezen dat ethisch beleggen kali samengaan met (een meer dan) aanvaardbaar rendement. Sommige, vooral Belgische fondsen, hebben uitstekende resultaten voorgelegd. Deze vaststelling, gecombineerd met een toenemend ethisch bewustzijn, kan andere financiële instcllingen aanmoedigen om het aanbod van ethische beleggingsfondsen en andere ethische producten uit te breiden. Deze evolutie kan alleen maar aangemoedigd worden!
NOTEN 1. Enkele niet-ethische effecten zijn Iiet smal1 coinpany effect, het diversificatie-effect, het survivorship bias fenomeen, het anticipalion effect ei] de lage dividei~duitkering.In Iiet punt 4.3 vaii "Etisch hcleggeii in de Benelux" worden deze effecten verder uitgewerkt.
REFERENTIES Blom, P.; Cocckelberg, D,; Duplat, J.-L. en Peeters, I-I., 1996, Verantwoord anders etliiscli beleggen binnen de Benelux, (Mys aiid Breescli, Gent), 111.
Een efficiënte besturing van liften
COSI'N Jan (K.U.Leuven, Handelsingenieur (1999)) De impact van liften op onze Westerse samenleving wordt voortdurend groter. In de steden worden er immers steeds hogere wolkenkrabbers gebouwd omwillen van de hoge grondprijzen. Hoe hoger de gebouwen, hoe groter de problemen om de mensen naar boven en naar beneden te krijgen. De reistijd van passagiers zal toenemen vermits de topsnelheid van liften begrensd wordt door beperkingen opgelegd door het menselijk lichaam. Steeds langere wacht- en reistijden zijn niet de enige en ook niet de voornaamste zorg in wolkenkrabbers. Problematisch is vooral het enorme ruimtebeslag van al die liftkokers. In de huidige generatie wolkenkrabbers nemen de liften al meer dan 10% van de ruimte in. Dit percentage neemt nog toe als de gebouwen hoger worden omdat er dan een bijkoinende nood aan express-liften naar de hoogste verdiepingen is. Een goed ontwerp van het liftsysteem en een goede liftbesturing kunnen ervoor zorgen dat er minder liften nodig zijn om een gevraagde dienstvrlening te leveren. De besparing die hieruit voortvloeit kan enorm zijn. Ze heeft zowel betrekking op de installatiekosten van het liftsysteem, op de ingenomen ruimte als op de energiekosten verbonden aan het liftsysteem. Analystische modellen van liften moeten omwille van de complexiteit van de materie grove vereenvoudingingen maken. Dit beperkt de bruikbaarheid van deze modellen en daarom is men overgeschakeld naar simulatie om de liftbewegingen te beschrijven. Vandaag is simulatie de enige manier om op een correcte manier de prestatie van complexe liftbesturingsalgoritmen met elkaar te vergelijken. In het eindwerk werd cen nieuw liftbesturingsalgoritme voorgesteld en vergeleken met bestaande algoritmen. Deze algoritmen - die beslissen aan welke lift een liftoproep wordt toegewezen - werden geprogrammeerd met behulp van de programmeertaal Visual Basic for Applications. De simulatie van de liftbewegingen die nodig zijn op de liftoproepen te beantwoorden werden geprogrammeerd in de simulatietaal SIMAN met behulp van het simulatiepakker ARENA. Alvorens verder in te gaan op de liftbesturingsalgoritmen is het nuttig even stil te staan bij de criteria die gebruikt worden oin deze algoritmen tegen mekaar a i te wegcn. Het is intuïtief antrekkelijk de gemiddelde wachttijd van een passagier als criterium te hanteren. De wachttijd heeft immers een grote psychologische invloed op de beoordeling van de dienstverlening door een individu. Men moct
echter oppassen met besturingssystemen die de wachttijd minimaliseren. Het is immers mogelijk dit te doen ten koste va de gemiddelde tijd die een passagier in het systeem doorbrengt (wachttijd transporttijd). Daarom werd er in het eindwerk voor geopteerd deze gemiddelde systeemtijd als criterium te hanteren, met als nevenbeperking dat de wachttijd niet te hoog (typisch 30 sec) mag oplopen. Daarbovenop komt nog de algemeen aanvaarde beperking dat het ongewenst is dat cen Ift van richting zou veranderen indien er zich passagiers in de lift bevinden. Het is mogelijk de J- of i-knoppen in dc lifthal te vervangen door een cijferklavier. De liftgebruiker kan in dat geval zijn eindbestemming aan het liftsysteem ineedelen vooraleer hij de lift betreden heeft. Hierdoor beschikt het liftbesturingsalgoritme over meer informatie wanner het over de toewijzing moet beslissing. Betere toewijzingen worden mogelijk door passagiers met éénzelfde bestemming in &énlift te groeperen. Hierdoor daalt het aantal stops per lift, wat de prestaties van het liftsysteem ten goede komt. Het nieuwe algoritme wijst de liftoproepen aan een lift toe zodanig dat de gemiddelde tijd die alle passagiers in het liftsysteem doorbrengen geminimaliseerd wordt. Deze doelfunctie is equivalent met het minimaliseren van de totale systeemtijd (= som van de systeemtijden van elke individuele passagier). Het algoritme verloopt als volgt: alle mogelijke combinaties van toewijzingen worden gegenereerd en hun totale systeemtijd wordt berekend. Tot slot wordt de toewijzingscombinatie met de kleinste totale systeemtijd gekozen. Het berekenen van de totale systeemtijd van een toewijzingscombinatie vergt een simulatie van de liftbewegingen via recnrsieve procedures in Visual Basic for Applications. Door de aard van het algoritme (het bevat een gedeelte enumeratie) is de toepasbaarheid ervan beperkt tot situaties met twee of drie liften. Bij meerdere liften en vele gelijktijdige liftoproepen zou de benodigde rekentijd voor het vinden van een optimale toewijzing zo groot worden dat het de prestaties van het liftsysteem schaadt. Niettemin is het algoritme waardevol, en dit om twee redenen: enerzijds, in een groot aantal gebouwen zijn de liften gegroepeerd per twee of drie liften; anderzijds geeft een simulatie met dit algoritme een bovengrens op de prestatie van het liftsysteem waarmee heuristieken met kleine rekentijd ktinnen vergeleken worden. De simulatie werd uitgevoerd voor een gebouw van 7 verdiepingen. Zowel de situatie met twee liften als met drie liften werden bestudeerd. Uit de experimeilten blijkt dat het nieuwe algoritme op basis van vroegtijdige kennis van de eindbestemming steeds beter presteert dan de bestaande algoritmen. Afhankelijk van de situatie (veel of weinig liftoproepen, verdeling van de liftoproepen over de verdiepingen) kan er een daling zijn tot 7% in de gemiddelde tijd die een passagier in het liftsysteem doorbrengt. Alleen tijdens de ochtendpiek wanneer alle mensen willen stijgen en bijna niepmand wenst de dalen, presteren de bestaande algoritmen even goed als het nieuws algoritme. Het eindwerk toont aan dat het te overwegen valt om de vertrouwde L- of ?-knoppen in de lifthal te vervangen door een cijferklavier. De verwachting is dat de grote kracht van algoritmen met vroegtijdige kennis van de eindbestemming ligt bij hoge gebouwen met een groot aantal liften. Verder onderzoek naar snelle algoritmen is dus zeker gerechtvaardigd.
+
1. Papers for publication should be sent in triplicate to: Tijdschrift voor Economie en Management Prof. Dr. i? Sercu p/a Mevr. A. Ronsmans Redactiesecretariaat Naamsestraat 6 9 - 3 0 0 0 Leuven, Belgium. Submission of a paper wil1 be held to imply that i t contains original unpublished work and is not being submitted for publication elsewhere. 2. Manuscripts should be typed double-spaced on one side of the paper only, and record the author's name clearly. 3 . Each paper should have an elaborate introduction and conclusion with summary. These have t o contain the reasons and relevance of the research reported, as wel1 as its main findings and their policy relevance.
4. Footnotes should be kept to a minimum and numbered consecutively. They are put at the end of the text.
5. The conventions for references are those of the European Economic Review. In the text, references t o publications should appear as follows: As argued by Goldfeld and Quandt ( 1 9 7 3 ) ... or: "Decision tables ... see Verhelst (1980). The author should make sure that there is a strict "one-to-one correspondence" between the names (years) in the text and those on the list. At the end of the manuscript (after any appendices), the complete references should be listed as: For monographs: Verhelst, M., 1 9 8 0 , De praktijk van beslissingstabellen (Kluwer, DeventerAntwerpen). For periodicals: Goldfeld, S. and Quandt, R.E., 1 9 7 3 , A Markov Model for Switching Regressions, Journal o f Econometrics 1, 3 - 1 5. For contributions t o collective works: Taylor, B., 1 9 7 0 , Financing Tables and the Future, in Taylor, B. ed., Investment Analysis and Portfolio Management, (St. Martin's Press, New York), 378-386. 6. Diagrams should be in a form suitable for immediate reproduction: 1 original drawn i n black ink on white paper and 2 photocopies. Care should be taken that lettering and symbols are of a comparable size. The drawings should not be inserted in the text and should be marked on the back with figure numbers, title of paper, and name of author. 7. Contributors are responsible for the correction of galley proofs. Corrections other than printer's errors may be charged t o the author. 3 copies are supplied free; additional copies are avalaible at cost if they are ordered when the proof is returned. 8 . Any manuscript which does not conform to the above instructions may be returned for the necessary revision before publication.
RICHTLIJNEN VOOR AUTEURS 1 . Te publiceren artikels moeten in drie exemplaren gezonden worden aan: Tijdschrift voor Economie en Management Prof. Dr. P. Sercu p/a Mevr. A. Ronsmans Redactiesecretariaat Naamsestraat 6 9 - 3 0 0 0 Leuven, Belgium. Elders gepubliceerde artikels worden niet aanvaard.
2. De teksten moeten getypt zijn op één kant van het papier, met dubbele interlinie met duidelijke vermelding van de auteursnaam. Artikels mogen in het Engels of het Nederlands gesteld zijn. Er kunnen nochtans slechts een beperkt aantal Engelstalige bijdragen per volume worden opgenomen. 3 . Elk artikel dient een uitgewerkte inleiding en besluit met samenvatting te bevatten, zodat de essentie en relevantie van de probleemstelling alsook de eigen bijdragen van de auteurs duidelijk, overzichtelijk en summier aan de lezer kunnen overkomen.
4. Voetnoten moeten tot een minimum herleid worden en opeenvolgend genummerd. Ze worden achteraan de eigenlijke tekst samengebracht.
5. De conventies van referenties zijn dezelfde als deze van de "European Economic Review". In de tekst voorkomende referenties moeten als volgt vermeld worden: "Zo betogen Goldfeld en Quandt ( 1 9 7 3 ) ... of "Deze beslissingstabellen zie Verhelst ( 1 9 8 0 ) . De lijst van referenties als volgt: Voor boeken: Verhelst, M., 1 9 8 0 , De praktijk van beslissingstabellen (Kluwer, DeventerAntwerpen). Voor periodieken: Goldfeld, S. and Quandt, R.E., 1 9 7 3 , A Markov Model for Switching Regressions, Journal o f Econometrics 1, 3 - 1 5. Voor verzameld werk: Taylor, B., 1 9 7 0 , Financing Tables and the Future, in Taylor, B. ed., Investment Analysis and Portfolio Management, (St. Martin's Press, New York), 3 7 8 - 3 8 6 . 6 . De in artikels voorkomende figuren moeten getekend zijn op een apart blad ( l origineel en 2 kopies) in zwarte inkt, op de achterzijde dient vermeld: naam auteur, titel artikel en figuur nummer. 7. De auteurs verbeteren de bandproeven. Extra-correcties op de drukproeven (d.w.z. verbeteringen die op afwijkingen van de ingezonden teksten neerkomen) brengen kosten met zich mee, ten laste van de auteurs.
8. De gebruikte spelling van de Nederlandstalige artikels is de voorkeurspelling. 9. Elk artikel dat aan de hierboven beschreven instructies niet beantwoordt wordt voor nodige herwerking teruggezonden.
