The Making Of The Statute Of The European System Of Central Banks ~ Chapter 8: Conclusions To Cluster II
This cluster has dealt with the articles determining the System’s operational functions and the division of labour between the ECB and the NCBs. There might be some tension between the ECB and the NCBs as regards this division of labour. It would increase the ECB’s position considerably, if it were conducting large part of the System’s operations with the banks and in the markets. A complete centralization (in all areas, e.g., open market operations, domestic facilities, foreign exchange operations, payment services, banknote issuance) would be unlikely, as this would be tantamount to merging all balance sheets, thus ending the federal character of the System. Complete decentralization, on the other hand, would not seem to be incompatible with a federal central bank system. Complete decentralization allows for concentration of activities with one of the central banks. Mixed systems are possible too. The centre could act as one of the central banks, offering all central bank services or conducting only a few types of transactions. To understand the way the Statute has been drafted we need to know what the drafters were aiming for when designing the Statute.
Some of the contours of the system though date from before the Delors Committee, because Germany had indicated (and France had supported this) that the system should be federal, i.e. with a clear role for the existing NCBs. Taking the Bundesbank as an example, NCBs could be expected to participate in the highest decision-making body and to have local operational functions. The Delors Committee did not go into details as regards the internal division of labour – this was not the core of their report. Nonetheless, already at this stage the Delors Committee agreed on pooling of reserves at the centre – some had in mind the final stage, others (among whom Delors and the French central bank) already the intermediate stage.
The Committee of Governors concentrated on the final stage: they agreed to pool a substantial part of the reserves and to make the management of the remaining reserves subject to central guidelines. Pooling would create a pool of directly available foreign reserve assets, lending credibility to the System’s exchange rate policy, and indirectly contributing to creating a ‘strong’ centre. A strong centre was considered desirable, because the System would probably have to operate – at least in its formative years – in a difficult environment, in a political sense. While the governors clearly had in mind that monetary policy would be implemented using the NCBs and that the System itself should decide on the division of labour within the System, they could not agree on the degree of decentralization (i.e. strong or very strong) and neither on which body should ultimately have the final say as regards the actual division of labour (the Executive Board or the Governing Council). Most of the governors wanted the Governing Council to decide, a small minority (mainly the Bundesbank) favored the Executive Board. Behind this was the fear that NCBs could undermine the singleness of monetary policy. The governors did agree though that in order to ensure the singleness of policy the Governing Council should be able to adopt guidelines, upon which the Executive Board could base instructions to NCBs, which NCBs were required to obey (see Art. 12.1, first and second paragraph and Art. 14.3).
The IGC quickly found a compromise, choosing for the strong option (and not the very strong option) and placing the decision in the hands of the Governing Council (and not the Executive Board). The IGC did not express a specific opinion on the desired degree of centralization or decentralization of the System’s operations. The IGC devoted some time to the topic of banknote issuance, basically to accommodate peculiarities of a few countries (relating to the issuing of banknotes by a few private banks and to the issuance of coins). Overall, the governors made a considerable mark on the final design of the System.
Taking the procedure used in chapter 5, we will first present a short summary of the articles of which we studied the genesis in chapter 7. We will highlight the specific interests at stake for the NCBs and/or the centre. After this we will analyse the checks and balances by focussing on the elements in the Statute which ‘protect’ the positions of the NCBs and which elements protect the role of the ECB (section 8.2.2). In section 8.2.3 we will show the ‘outcome’ in practice, i.e. how have the operational functions been allocated at the start of the ESCB in 1999. To put this in perspective we will ask ourselves which other ‘models’ are possible (section 8.2.4). This may shed light on the possible future trends in the allocation of operational powers. Finally, we use the categorization of checks and balances developed in chapter 2, which allows us to detect possible unevenness in the checks and balances between the ECB and the NCBs. This allows us to present some suggestions for improving the checks and balances relating to the internal division of labour (section 8.3).
8.2 Checks and Balances between the ECB and the NCBs
8.2.1 – Short reviews
8.2.2 – Factors influencing the division of operational powers
8.2.3 – Actual situation
8.2.4 – Alternative models and possible future trends
8.2.1 Short reviews (with emphasis on checks and balances)
Supervision (Article 3.3- and 25-ESCB):
Discussion on this article on prudential supervision and financial stability was complicated because the existing national arrangements differed. Some central banks were responsible for supervision; others (in fact most of them) were only ‘involved’ through cooperation with or membership of their governor of a separate supervisory body. In practical terms, though, all central banks, except the Danish, were actively involved in the supervisory process. Financial stability, i.e. the stability of the financial system, was increasingly seen as an area calling for active central bank involvement, though this had as yet not been reflected in the mandates of the NCBs.
The reluctance of the German central bank to be made responsible for supervision combined with the reluctance of other NCBs/countries to share their supervisory responsibilities, which had always been a closely guarded functional area, with other countries/central banks. This reluctance was also felt vis-à-vis the ECB, which was a new potentially powerful institution, which might start encroaching on the supervisors’ territory. In retrospect, there was no willingness to make optimal use of the opportunity to create synergy in the field of supervision; the national roles could adequately have been safeguarded by explicitly putting down on paper the (minimum) involvement of NCBs in supervision, e.g. in the area of on-site inspections. However, what happened was that the NCBs, and also the Finance Ministries, seeing the ECB as a potential supervisory competitor, marginalised the supervisory role of the System and the ECB, though a potential role was not excluded.
This brings us to the following general observation leading to two suggestions for improvement. Protecting financial stability in the euro area requires an effective network of cooperation and information sharing between national supervisory authorities. What is true at the national level, i.e. that there is potential synergy between being monetary authority and being involved in supervisory tasks, is also true at the European level. At the same time, the problems Germany had (and has) with a possible conflict of interest between monetary and supervision at the European level should be coped with. This could be achieved by copying the German-style involvement of the ECB in supervision, through a Treaty article explicitly obliging central banks and supervisors to inform the ECB of relevant developments (first improvement) and by making the ECB for instance co-supervisor for large foreign banks or large bank holdings (second improvement). The tasks of the ECB need to be spelt out quite clearly. The requirement to inform and be informed would also allow the ECB to make most of its role as contributor to financial stability. The ECB is uniquely placed to oversee financial stability issues on a daily basis for two reasons: it has a European scope and it has daily contacts with the financial markets. This makes a difference with other supervisors: national supervisors only have a national scope and European regulatory bodies only meet infrequently. The institutional balance between the NCBs and the ECB, if considered important in this respect, would not be disturbed, because the ECB would only be co-supervising with the relevant national supervisors (i.e. the ECB would not substitute for supervision by national supervisors), but at the same time it would enable the system to acquit itself more effectively of one of its tasks.  The appointment as co-supervisor could be enacted through activation of Art. 25.2, which could possibly be confined to specific supervisory issues. The Executive Board would be informed on a continuous basis and the Governing Council would be informed regularly, or when deemed necessary, while regular reporting should also take place in relevant European fora.
Collection of statistics (Article 5-ESCB):
Though the collection of statistical data does not have the mystique of monetary policy making, it involves large contingents of people to collect, analyse and report these data, which are indispensable for informed policy-making. The drafters of the Statute introduced a logical split between policy-making by the central body (e.g. cooperating with other bodies, contributing to harmonization) and decentralized execution (NCBs collect ‘to the extent possible’). Decentralized execution makes sense from the point of view of effectiveness and efficiency (i.a. language problems and national peculiarities), but it increases the distance between the institution asking for information (basically the ECB) and the persons/agents carrying the reporting burden, thus possibly introducing a bias towards ‘over-asking’. The ESCB now applies a cost-benefit analysis to each new information demand, but it would not have been wrong to state the ‘obvious’ (‘thou shalt not unnecessarily burden reporting agencies’) in the Statute, following the example of the Federal Reserve. The article only relates to the collection and not the reporting of information. Practice has it that the ECB only reports euro area aggregates, while NCBs publish relevant national data. NCBs should not publish sensitive national monetary data (e.g. a national money growth figure) before the aggregate is published, as this would make monetary policy making more difficult.
International cooperation (Article 6-ESCB):
It was considered very important that the System would represent itself to the outside world as a unity. Art. 6 allows the Governing Council to decide on the System’s representation, but also on the System’s position or common language, e.g. in case the System is represented by both the ECB and NCB representatives. A binding approach is possible in areas where the System is competent. It is not possible where the System has an interest, but lacks authority, because then NCBs cannot be bound. Then there can only be voluntary coordination of opinions. Coordination strengthens the ‘expert’ opinion of the System. In these areas the president of the ECB should make clear whether he speaks for the Board or the System. Art. 6 does not prejudge the way the System should be represented in international meetings. It allows NCBs and ECB to participate alongside each other, which makes sense anyhow because most gatherings also relate to non-System responsibilities of NCBs. In theory, the Governing Council could decide that the System is represented by an Executive Board member together with a representative of the NCBs with a special kind of expertise, to the analogy of the Federal Reserve, which in international meetings usually is represented both by someone from the Board and from New York.
It should be noted that were NCBs to lose their international contacts, their presidents would lose influence in the Governing Council and the quality of the monetary policy debate could stand to lose too, because well-informed people are better able to take the best possible decision than non-informed people, especially when they receive their information from a plurality of sources (and not only from one staff presentation). Here we see a reflection of the general idea that the division of operational competences is of direct relevance for the quality of monetary policy making.
Decentralization of operations (Article 12.1c-ESCB):
There is an analogy with Art. 6: if NCBs would lose their contact with the markets their presidents would be less well informed and would contribute less to the discussions in the Governing Council. The article can best be understood by looking at its origin: the first draft showed a relatively centralized system. The Germans had wanted this to ensure uniformity; the Dutch wanted this too, because they thought this to be supportive of the independence of the System. The Delors Report had recommended the centre to have its own balance sheet (read: holding foreign reserves), making it already more than the Board of Governors, which is a decision-maker, regulator and overseer, but does not execute operations. However, the French had no wish to centralize the domestic monetary assets. To this end they successfully introduced a strong bias towards decentralized implementation. They argued this would not undermine the unity of the System, because the NCBs should be under very strict operating guidelines. This was supported by most other central banks. Arguments used in favour of decentralized operations were: (1) the subsidiarity principle (though it was used out of context – see box in section II.3 of Art. 12.1c in chapter 7 –, it had strong appeal and was an effective argument), (2) cost efficiency (existing infrastructures could be used), (3) the argument that several financial centres should co-exist (the alternative being the development of a European equivalent of New York with shrunken provincial financial centres in the regions, while the creation of one financial centre would probably also give more clout and political influence to the financial sector), (4) continuity – it would allow most commercial banks to continue to participate in the money market operations of the central bank. And more in general it enhances the federal character of the system. Apart from this, there are synergies when monetary authorities are also involved in supervision (in terms of knowledge of the financial markets and institutions, which would of course dry up when the monetary function becomes non-existent). The IGC was asked to decide on the exact formulation of the bias favouring decentralization (basically a choice between a very strong and an even stronger bias). The IGC took a middle position, but – and this is at least as important – it accepted that the precise balance could be determined in practice by the System itself.
The ECB would seem to suffer less from an informational disadvantage, because it is informed daily by the NCBs of market developments, while it is also responsible for ensuring that monetary policy is implemented and for the conduct of possible foreign exchange interventions. This implies the ECB is in regular contact with the markets and is well-informed.
NCBs and the System (Article 14.3 and 14.4-ESCB):
It was not contested that NCBs would be more than mere agencies of the ESCB. They were allowed to perform independent tasks, thus preserving their domestic profile and strengthening the position of the System (because it allows for synergies), and the position of those governors with important non-System tasks. It does not seem to harm the ECB, because it does not take away tasks of the ECB, though it affects the balance between the ECB and the NCBs to the extent that it strengthens the NCB’s identity and their independent right to exist. The attribution of non-System tasks to an NCB might though, under circumstances, give the national political authorities unwanted leverage over the NCB’s board (see also Art. 7, footnote 12). Art. 14.4 was balanced by a text drafted by Tietmeyer entailing complete subjugation of NCBs to the objectives of the System, as evidenced by the fact that the Governing Council may forbid functions which are detrimental to, or even just interfere with, the System’s objectives and tasks. (Because such a fundamental decision should not be taken lightly, a two-thirds majority is required.) The formulation of Art. 14.3 could be said to transcend Art. 14.4: it has a high federal content. The obligation for NCBs to provide the Governing Council with information provides a form of accountability vis-à-vis the Governing Council.
It follows that non-System tasks should also not lead to the possibility for political authorities to dismiss NCB board members for failing to please the authorities in fulfilling these non-System functions, as this could be misused to dismiss a board member for monetary policy reasons. Of course, the general clause of being unfit to fulfil a board function would still apply, while alternatively non-System tasks could be taken out of the NCB by the national authorities.
Banknote issuance (Article 16-ESCB):
The article looks innocent, but its innocence hides a very long story. For our purpose the following is relevant: policy-decisions were laid in the hands of the Governing Council. Banknote issuance is possible both by the ECB and the NCBs. The drafters did not distinguish between legal and physical issuing. Elsewhere, however, Art. 12.1c leaves no doubt that the physical part (the distribution) should be carried out by the NCBs. Art. 12.1c does not give guidance as to who should be the legal issuer, i.e. who should carry the banknotes as liabilities on its balance sheet. After a long discussion the Governing Council would decide to separate the physical and legal aspect of issuing by allocating the banknotes in circulation over the ECB (fixed share) and the NCBs (using NCBs’ shares in the capital key), thus supporting the assumption that banknotes by different euro area NCBs are fungible. Sensitivities ran high, because NCBs view banknote issuance as a core NCB function – losing this would take away a very visible function, a function that makes their existence self-evident to the general public. Sharing the note issuance with the ECB took away a feature of uniqueness and indispensability of the NCBs. Distribution was laid with the NCBs. Production does not seem to be an exclusive System function, because some NCBs have their notes printed by private sector companies (of course, the System is responsible for their design and safety features and for having sufficient banknotes printed and in stock). In the area of production the ECB has taken the role of co-ordinator (this could have been an example for specialization among the NCBs, but in practice they are not keen in allowing the lead to a colleague NCB).
Monetary functions and operations (Article 17 – 24-ESCB):
These articles describe the System’s monetary instruments and other ‘allowed’ functions in generic terms. Details of the monetary instruments and procedures are provided in the General Documentation, as approved by the Governing Council ex Art. 18.2-ESCB. A backbone for the decentralized execution of monetary policy is TARGET, a real-time gross settlement payment system connecting all NCBs and accessible for all financial institutions. TARGET allows for a continuous uniform money market rate throughout the euro area.
The Statute allows both the ECB and NCBs to perform the execution of the System’s monetary and payment system functions. The first versions of the draft Statute prepared by the Committee of Governors referred to the ‘System’ as operator. When the legal experts of the NCBs advised not to give legal personality to the System, but only to its components, they also advised to attribute operational functions to both the ECB and the NCBs, because this would allow for development in the operational methods of the System without the need to amend the Statute. This was meant as an enabling mechanism and was not meant to prejudge the outcome. Nonetheless, it put the ECB completely ex aequo with the NCBs, which was not what the governors initially had had in mind. In theory, the chosen formulation allows both for complete centralization as complete decentralization. The only guiding principle is laid down in Art. 12.1c, favouring decentralized operations. Although that principle is strongly worded, two uncertainties remain: first, the words ‘to the extent deemed possible and appropriate’ need to be interpreted, and second, though the interpreter is known (i.e. the Governing Council), its composition is not. Indeed the Governing Council’s composition could be changed, though this requires an amendment of the Treaty. A reduced presence of the NCB presidents in the Governing Council (or reduced voting rights) could lead to an interpretation more in favour of centralization. The opposite, i.e. a reduced role for the Executive Board, seems unlikely. That is, the position of the Executive Board seems more secure. In other words, a change in the governance structure of the ESCB (e.g. the composition of the Governing Council) may affect the checks and balances also in the operational sphere. The System’s instrumental independence is based on the Statute providing an extended set of instruments, which can be used without ministerial involvement.
The System’s financial provisions (Article Art. 26 – 33-ESCB):
The intra-System financial relations do not give rise to one-sided dependencies. The centre does not control the NCB budgets, neither is the opposite true: decisions on the ECB’s budget are not taken by the NCBs (the shareholders), but by the Governing Council, in which the Executive Board is represented with a (large) minority (six out of fifteen votes). In fact, if budgetary decisions are prepared and supported by all Board members, they only need the support of a few governors to have their proposals approved. On the other hand, until now most Governing Council decisions have been taken by consensus, and one would expect that, except in urgent matters or monetary policy decisions, vote taking will be the exception. These budget decisions allow the Governing Council to decide on the staff size of the ECB.
Factors influencing the division of operational powers
We have condensed the key elements of the articles of cluster II in the previous pages. In this part we will try to answer the question whether the design is balanced as to the relative operational roles for the NCBs and the centre (ECB). If one party is considerably less protected than the other we have a system which in the long run is potentially unbalanced. A difference with the full definition of checks and balances developed in chapter 2 is that the ECB and the NCBs cannot decide themselves to encroach on the area of the other – this requires a decision by the Governing Council. Nonetheless, the threat is there and furthermore the ECB and NCBs need each other and there is some kind of accountability in the form of Governing Council approval for the ECB’s budget and the requirement for NCBs to provide the ECB with any information the ECB considers necessary for it to ensure compliance with its guidelines and instructions.
Below we make a short inventory of which articles, or elements thereof, offer protection for the operational position of the ECB and NCBs respectively. We then confront this with a description of how the operational powers have been attributed in practice, basing ourselves on the General Documentation. We compare this ‘ESCB’ model with that of the Federal Reserve and the Bundesbank. This will allow us to draw some conclusions as to the stability of the design and possible future trends.
Table 8-1: Inventory of protecting factors
Factors protecting the operational position of the ECB
- The ECB has initially been endowed with euro 50 billion euro in foreign assets. Further transfers of reserves from the NCBs to the ECB are possible, according to a procedure involving the Council of Ministers (Art. 30).
- Foreign exchange interventions are the prerogative of the ECB. This follows from Art. 30.1, last sentence, and Art. 31.2. In this area decentralization is an option, but not required and also not practical.
- The article on international cooperation (Art. 6) allows the Governing Council to give the function of external representation completely to the ECB (read: Executive Board). The other extreme, i.e. placing this responsibility completely with NCBs, seems unimaginable, also because Art. 6.2 seems to suggest a natural right for the ECB. It would have to be mandated by the Governing Council (which could be a broad mandate). Allowing the Executive Board the monopoly on international economic and monetary information would substantially increase their leverage over Governing Council meetings.
Factors protecting the operational position of the NCBs
- In the statistical area the NCBs would seem to have an indisputed, strong legal base for continuation of their role as collector of necessary data from local reporting agents (Art. 5).
- NCBs are allowed to continue to perform transactions in fulfilment of their obligations towards international organizations (Art. 31.1), e.g. transactions with the IMF on behalf of their Member State for which they act as ‘fiscal agent’. This presupposes that some reserves will remain with the NCBs. The situation would change when the EU would become member of the IMF.
- The fact that NCBs may perform other non-System functions, e.g. in the area of supervision, gives them an edge over the ECB.
- If there is financial leverage from one group over the other, then it is from the NCBs over the ECB. The Governing Council, in which the governors form a majority, has to approve the ECB’s budget, and decides indirectly on issues like the size of the ECB staff and whether or not to set up a dealing room at the ECB.
- Pièce de resistance however for the NCBs is Art. 12.1c. This article introduces a strong bias towards decentralized operations. The article covers each of the System’s tasks, like open market operations, banknote distribution, payment system facilities, fiscal agent functions (but not so much the holding of the ECB’s own foreign reserves and their use for interventions). The article, however, is a procedural one: it does not give basic rights to the NCBs, but it provides a guiding principle for the decision-making bodies of the ECB. Again, the fact the NCB governors have a majority in the Governing Council is relevant. Until now it has not been deemed appropriate to deviate from this principle – see next sub-section.
We did not mention the issue of banknotes issuance. The practical aspects (distribution) fall under Art. 12.1c. The decisions taken as regards legal issuership have disconnected distribution from legal issuership. As long as most of the System’s assets remain on the NCBs’ balance sheets, it would seem natural that banknotes appear on the NCBs’ balance sheets as well, apart from a symbolic share for the ECB. Though symbolic, it means part of the System’s seigniorage is generated on the balance sheet of the ECB. That does not change the fact that the ECB’s expenditures need Governing Council approval. Furthermore, the ECB’s net profits flow to the NCBs. By allowing both the ECB and the NCBs to act as legal issuer, the drafters of the Statute have created in theory the possibility of a complete substitution of the NCBs by the ECB.
From the list we conclude that the ECB is especially ‘strong’ in the international area. It would seem difficult for the ECB to take away operational tasks from the NCBs, as this is determined by the Governing Council. The NCBs seem to have fewer basic rights than the ECB. Art. 12.1c does not provide a right, but only a direction. Nonetheless, the direction is loud and clear, viz. execution of tasks as much as possible through local NCBs.
We will now proceed with showing the actual outcome, basing ourselves largely on the ECB’s General Documentation on ESCB monetary policy instruments and procedures, approved by the Governing Council in September 1998.
Below we present a table summarizing the operational roles of the ECB and the NCBs.
Table 8-2: Summary table on division of labour with respect to ESCB operations and functions
———————————————————————- ECB NCBs
– Minimum reserve/payment accounts – x
– Standing facilities – x
– Open market operations:
– Refinancing operations – x
– Fine-tuning operations
– reverse transactions – x
– outright x x
– Structural operations
– reverse transactions – x
– outright x x
– Holding foreign reserves x x
– Managing ECB’s foreign reserves – x
– Managing non-pooled foreign reserves – x
– Forex interventions x –
– Managing euro denominated assets –  x
– Legal issuer euro banknotes x x
– Distribution and intake banknotes – x
– Collecting statistics at source – x
– Providing payment services –  x
– Payment systems oversight x x
– International cooperation x  x
– Member of BIS x x
– Fiscal agent – x
– Fiscal agent for national
– authorities in IMF –  x
– Represented at the IMF (x) –
The table shows an overwhelming majority of functions being performed by the NCBs. At the same time, all operations in the monetary field are conducted under strict rules determined by the General Documentation. NCBs have more room for manoeuvre when managing domestic (euro) asset portfolio’s, though even then NCBs have to respect the maximum size of operations, above which they need ex ante approval by the ECB, to prevent interference with monetary policy objectives. Management of the ECB’s foreign reserve assets has for the time being been delegated to NCBs, but this could be reversed. An area where NCBs are dominant is the provision of financial services: this includes the offering of payment accounts, payment services and securities handling. Also coin and currency distribution is an NCB affair.
Not mentioned in the table are the non-system functions, such as actual banking supervision. The actual division of labour as presented in the table clearly reflects the drafters’ intention, i.e. that the System’s policies (to be decided centrally) should be executed as much as possible by the NCBs.
Alternative models and possible future trends
One could ask the question whether the present division of labour is stable, or whether a movement away from decentralization is likely to occur. To investigate this, we look into developments that took place in the FRS and the other major country with a federal central bank system, Germany. There are no studies in this respect focussing explicitly on this question. We begin with a factual description of the present situation in the US and Germany.
In case of the Fed the Board of Governors would only have a cross in table 8.2 on the lines ‘payment systems oversight’ (and then only in the regulatory sense), ‘international cooperation’ and ‘member of BIS’. The external relations are completely in the domain of the Board, except for the operational side, which rests completely with the New York Fed as the yearly elected manager of the SOMA. The operational tasks rest with the FRBs, of which the following are localized in New York: open market operations, foreign reserve related activities (there would be no line on non-pooled reserves), managing domestic currency denominated assets, fiscal agent in IMF and member of the BIS. Therefore the ECB looks more like a central bank than the Board of Governors, which is a regulatory organ, though of course very important.
In case of the Bundesbank the table would be more complex than for the Fed. The Bundesbank Head Office deals with banks which have a federal significance. To be more precise: the Head Office handles in particular ‘transactions with the Federal Administration and its Special Funds, transactions with credit institutions which have central functions for the whole of the Federal territory, foreign exchange transactions and other transactions with foreign countries, and open-market operations.’  On the other hand, the following business is reserved for the Landeszentralbanks (LZBs): ‘transactions with the Land and public administrations and transactions with credit institutions within their area of competence, except transactions with credit institutions which have central functions for the whole of the Federal territory.’  In doing so, ‘the LZBs are subject to the decisions taken by the Central Bank Council in matters of credit and monetary policy and to the general directives issued by the Central Bank Council, which may be supplemented in individual cases – if special circumstances should warrant it – by particular instructions.’ Then the study we quote from, written by a former LZB president, refers to the principle of decentralization: ‘Even in the case of decisions, instructions and measures of the central authority that are generally binding on the whole system the principle of decentralization is indirectly safeguarded by the fact that the Landeszentralbanken participate through their Presidents in the formulation of policy by the Central Bank Council.’ Here we see a much larger degree of centralization than in the US, but still a backbone of operational responsibilities for the local central banks. But again the international side is completely in the hands of the centre.
We conclude from the above that a decentralized system is not inherently unstable, it is workable and has proven to be effective. This does not preclude changes from occurring. Comparing the ECB with the Fed and the Bundesbank we see three areas in which changes might occur, all of them going in the direction of more centralization. We will treat them successively, after which we will make a few general observations relating to the longer term.
First, one could think of changes entailing a further or even complete centralization of external contacts and business. Thus the ECB would move in the direction of the Fed and the Bundesbank. However, what makes for an important difference with both the Fed and the Bundesbank is the existence of other, non-System functions of the euro area NCBs. This creates an independent reason for them to stay internationally active. Not being able to participate actively in international circles and fora would reduce the NCBs’ influence vis-à-vis the ECB.
A second development could be more centralization of open market operations. This could take two forms: (i) banks could be allowed to open accounts with the ECB. This implies the ECB would have to open accounts for thousands of European banks, including for very small ones. Though practically not impossible, it is an unlikely development, because it offers little benefits over the present system and it would violate the decentralization principle. A disadvantage for NCBs is that they would lose contacts with important players in national payment systems and they would lose the synergies resulting from their being both monetary authority and supervisor (or co-supervisor). (ii) An alternative could be the introduction of a primary dealer system (or clearing bank system), the essence of which is that it is two- or more layered (liquidity would reach banks via a limited number of specialized banks or financial houses that deal directly with the central bank). In case of a primary dealer system one would need one point of communication with the primary dealers. There will be problems both with respect to the choice of the primary dealers as the central point of entry. As regards the primary dealer system, if one would opt for one primary dealer per country, the selection of them will be fraught with ‘political sensitivities’, e.g. should the biggest bank from a small country with low transaction volumes be included and much larger banks, but not being the largest in their country, be excluded?
The logic alternative is working with those primary dealers which can guarantee large volumes. This option would probably lead to a concentration of open market operations in one (or a few) financial centres. This leads to severe political problems, as the other financial centres would stand to lose, if not in terms of direct business, then in prestige (translating in indirect loss of business). As regards the single point of entry, one could decide to periodically elect one central bank that is authorized to manage the system’s open market account; compare the New York Fed which manages the Fed’s SOMA. Such a function could rotate, but that is highly impractical. Therefore, such a procedure will in practice lead to specialization. Art. 12.1c refers to the ECB being obliged to take recourse to the extent possible and appropriate to the NCBs, this implies the ECB cannot ‘impose’ specialization in the operational field – therefore, all NCBs would have to agree. (If the NCBs would agree to interpret Art. 12.1c in such a flexible manner, they could opt for the ECB as well.)
In a primary dealer system the number of regular contacts of the System with banks would shrink, making for less synergy, though banks would still manage their reserve accounts with their local NCB. This is a disadvantage for central banks that appreciate direct contacts with their banks, inter alia because they usually are prudential supervisor, or at least are actively involved in supervisory work. Furthermore, practical changes would become necessary: for instance, banks now also use local collateral (so-called tier-two assets), e.g. bank loans, in refinancing operations with their central bank. It remains to be seen whether primary dealers would accept that kind of collateral too. We conclude that while a primary dealer system would technically be conceivable too in the euro area, there is no technical need to change the present system; moreover, a change to a primary dealer system would trigger a number of issues difficult to solve and would increase the distance between the ESCB and many banks. We conclude from the above that centralization or specialization in this area is unlikely to be proposed by the ESCB. If nonetheless centralization at the ECB were to happen, it would reduce the influence of NCBs and increase the influence of the ECB. Specialization would probably lead to intensified contacts between the ECB and the NCB responsible for (most) OMOs, like in the U.S. between the Board and the New York Fed. The ECB would probably gain from this development in terms of influence over the System’s operations and decisions.
A third development one could think of is the general abolishment of the prohibition of remote access to central bank credit (see description of Art. 17). This would have a cascade of consequences, because a commercial bank would want to hold its payment account which serves as we have seen also as its minimum reserve account at the central bank from which it receives its central bank liquidity (which is credited to its payment account). Therefore, this could affect almost all business areas of the central banks, depending on how many commercial banks would opt for another central bank. In this respect is relevant that domestic transactions between banks which are handled through their central bank RTGS systems would become cross-border payments, if one of the banks changed its central bank account to another central bank. Cross-border transactions through TARGET are more expensive than domestic RTGS transactions. It is possible that only a few commercial banks with large international transactions volumes would actively search for the cheapest payment services, but the system would get more complicated to handle. At the same time we see that the central banks themselves strive for as much harmonized payment services as possible. We contend that a development in the direction of harmonization is more likely than towards centralization (towards the centre) or specialization (towards one or a few NCBs). In this process of harmonization the ECB could play a proactive role. Such a role is within its competence. One could even say the centre has a duty to enable and facilitate decentralization of operations. Remote access would probably lead to some specialization, but the difference with the specialization in the preceding paragraph is that there would not be a primary dealer system, but all banks would still receive their liquidity directly from a central bank.
We have now dealt with possible developments towards more centralization. Theoretically, the opposite, i.e. a development towards less centralization, is possible too, but that is quite unlikely in view of the already very high degree of decentralization.
The above relates in first instance to the short-term and to developments which can take place within the context of the present Statute. In the short-run we do not think it is likely that major changes will take place. In the longer run though major changes could happen. Major changes have occurred in the Federal Reserve Act and in the set up of the Bundesbank. However, in these cases major changes were triggered by major external developments, as we will see below. The number of Landeszentralbanken (which threatened to increase from eleven to sixteen at the occasion of the German unification) was instead reduced to nine, by creating state-overlapping State central banks. In the United States major changes occurred in 1933 and 1935, after the dramatic event of the Depression. FRBs were forbidden to make their own arrangements with foreign central banks. Open market operations were completely centralized in New York (this process had already started in 1922 – see appendix 1 below) and they came under complete control of the FOMC. At the same time, the Board of Governors, formerly only a supervisor over the FOMC, became member of the FOMC and gained a majority of the votes. This is sometimes seen as a major step towards centralization. However, it is more appropriate to describe this development as a reaction to the increased importance of open market operations, which had not been foreseen by the drafters of the FRA, and which thus called for ‘reparation’ of the FRA to restore the original balance. The reparation had become urgent, because the Fed had not functioned effectively in the years before and during the Depression. Making the appointment of FRB presidents by their boards of directors subject to approval by the Board of Governors was part of this reform. The Federal Reserve remained nonetheless still a federal system. Its federal character is engrained in the Federal Reserve Act, the first words of which are: ‘To provide for the establishment of Federal reserve banks’. The federal character of the ESCB is also reflected in the first article of the Statute of ESCB and the ECB: ‘The ESCB and the ECB shall be established …..’. In contrast the Bundesbank had a less federal character: the State central banks are branches of the Bundesbank and the Head Office conducts open market operations directly with major banks.
Therefore, we see that major external shocks, both political (re-unification) and market-related (growing importance of OMOs), can lead to changes in the structure of federally designed central bank systems. In the cases mentioned above these changes, though being major, did not undermine the federal character of the Systems; in fact, one could say they were meant to prevent or correct a weakening of the centre. Nonetheless, in the long run major elements of the system are not excluded. In case such changes are proposed, one should carefully consider what the motives are for doing so, before one allows for a change in the balance between central and decentral elements, one side-risk being that a scaling back of the operational roles of NCBs will in the end undermine their presence in the Governing Council. The operational knowledge of the centre would seem to be adequately safeguarded by the fact that the centre prepares all implementation decisions and the fact that the centre is responsible for foreign exchange management and policy. To make an analogy with the Fed, the Board of Governors never seems insufficiently informed about the markets, although it does not carry out market operations itself. The presence of NCB governors in the Governing Council, in other words the federal composition thereof, also strengthens one of the other anchors of the ESCB, i.e. its independence as a monetary policy maker. Furthermore, direct operational knowledge benefits the discussion on monetary policy. Potential efficiency gains, when central bank operations would be carried out by the ECB, would have to be weighed against advantages of continued decentralized operations. In this respect we refer to Goodfriend (2000), who concludes that a regional presence facilitates surveillance of the economy and helps a central bank to communicate with the public. Competition among regional central banks stimulates innovative thinking on policy, research and operational questions. We add that many central bank tasks also benefit from proximity to the banks, like payment services, standing facilities, banknote distribution, supporting supervisory activities. This leaves uncontested that cost considerations could enter the equation. It will be clear that the combination of decentralized implementation and the transfer of profits to the national Treasuries does not make for cost awareness being given high priority by the System. At the same time, more cost efficiency does not necessarily mean centralization of monetary operations. Harmonization and sharing of RTGS systems would reduce costs though.
8.3 Overview of Checks and Balances between Executive Board and Governors and Possible Improvements
In this section we apply the system of checks and balances of chapter 2 on the European System of Central Banks (ESCB), seen as a federally designed system in which separate bodies (the ECB and NCBs) are united within an overarching structure, in which each body, while maintaining its individuality, shares in the operational tasks of the system. To this end we divide the articles of this cluster over the categories of checks and balances mentioned in chapter 2. We follow the presentation used in chapter 5.4, but this time related to the ‘internal’ checks and balances between the ECB and the NCBs. There are at least fifteen articles and sub-articles containing ‘internal’ (ECB-NCB) checks and balances (not counting chapter IV).
Table 8-3: Overview ‘internal’ checks and balances (between ECB and NCBs)
(a1) Checks and balances protecting the prerogatives of the ECB:
- Art.6; 9.2; 30.1; 30.5
(a2) Checks and balances protecting the prerogatives of the NCBs:
- Art.3.3; 5; 12.1c; 14.4; 31.1; 32; 33;
- Controlling (or blocking) mechanisms:
- Art.6; 14.3; 14.4; 31.2
- Consultation mechanisms:
- (see Art. 12.3 of cluster III on the role of ESCB committees)
- Accountability mechanisms:
- Art.12.3; 14.3 (last sentence); 26
- Checks and balances allowing for intertemporal flexibility
- Art.5.2; 6; 12.1c; 16; 17-24; 25.2
Explanation: Categories (a1) and (a2) contain the articles defining operational tasks which cannot be taken away. Line a1 contains the task for the ECB of seeing to it that the System performs its tasks and of being in control of the pooled foreign reserves, while the ECB is also allowed to accept invitations to participate in (the capital of) international monetary institutions. Line a2 contains the rights of the NCBs, though some of the articles mentioned relate to non-System tasks (Art. 3.3 and 14.4), while Art. 5 and 12.1c contain the decentralization principle. Art. 12.1c applies to the typical central bank functions contained in Art. 16 (issuance and distribution of banknotes) and Art. 17-24 (monetary functions and procedures). Art. 31-33 protect the NCBs’ financial rights.
The blocking mechanisms (category b) mostly refer to possibilities for the centre (ECB, but read: Governing Council) to impose restrictions on or guidelines for NCBs (Art. 6; 14.3; 14.4; 31). There are no examples of blocking power of NCBs vis-à-vis the ECB. Category (c) is empty. Indeed, formally the ECB and NCBs do not consult each other, but in practice their officials consult almost continuously, because they ‘meet’ in (advisory and technical) ESCB committees and in the Governing Council. The accountability mechanisms (category d) relate to information requirements. The ECB (Governing Council) may require from the NCBs all information necessary for it to be able to check whether the NCBs act according to and within the guidelines and instructions of the ECB (Art. 14.3). One would assume though that this information requirement also applies to the ECB. The annual accounts of the ECB, but also its annual budget (and with it the headcount), have to be presented to and be approved by the Governing Council.
A number of articles allows for flexibility (category e). They open the possibility to share out operational tasks over the ECB and NCBs and the possibility to give the ECB specific tasks in the supervisory area.
Table 8-3 captures the checks and balances of the functioning of the federally designed ESCB, these checks and balances determine the relative roles of the NCBs and the ECB within a flexible framework. They determine the inalienable rights, the way the components of the System depend on each other and the way their powers are limited (e.g. by guidelines) and the elements of flexibility.
Table 8-3 shows two specific features, see also diagram 2 in chapter 12. First, the table shows that the number of articles allowing for flexibility (category e) is remarkably high, which allows for a high adaptability but also creates a relatively high uncertainty as to the division of labour. In fact, category (e) contains the System’s core operational tasks (Art. 16, 17-24). Strictly speaking both extremes are possible, i.e. that only the NCBs perform these tasks or only the ECB. Neither is ensured of a continued operational role. The division of labour is relatively undetermined.
Second, the ECB has relatively few basic operational prerogatives, its edge seems to lie in the external area (international representation and being responsible for managing the pooled reserves). In contrast, category (a2), which is supportive of the position of NCBs, seems very well represented. However line a2 basically contains no more than procedural, financial and non-System related articles. We also note that Art. 5.2 and 12, though containing strongly formulated decentralization principles, do not give absolute certainty.
At the same time the NCBs are under pressure to reduce their size. Political authorities are taking a new look at their NCBs, because the NCBs stopped being extensions of the government or, where they were independent or important for anchoring low inflationary expectations or stabilizing the exchange rate, they lost importance to the ECB. Furthermore, a number of European central banks are relatively quite large consuming a lot of resources. Especially the Banque de France and the Bundesbank, which have a high number of local branches, have announced, or are engaged in, plans to reduce size. The (research) output of the NCBs is also scrutinized by studies, like the one from Eijffinger (2002).
Another development is the EU’s enlargement, implying increased importance of the ECB because not all new NCBs will perform all tasks (for instance some NCBs are too small to have their own RTGS (payments system) platform), and because the ECB will increasingly become the System’s external face, as the System will in the end come to encompass twenty-five or more central banks, of which none has the standing or unique position of New York. This pleads for less open-endedness of the division of labour between the ECB and the NCBs, and a higher and more visible cost-awareness of the System’s efficiency. In fact, this will be mutually beneficial, because NCBs which are sure of a minimum operational role are probably more co-operative as regards the System’s efficiency.
From the foregoing it follows that there are several areas in which the checks and balances in the operational field could be improved. In section 8.3.2 we will mention several improvements, some of which would require a change in the Treaty or the Statute; other improvements have a more practical nature and can be implemented without a Treaty change. Finally we mention a few suggestions which would allow the ECB to play a more meaningful role in the preservation of financial stability, while preserving supervisory responsibility as a national affair (though implying sharing information or even tasks with the ECB).
Apart from this, major improvements would not seem necessary, because we have concluded in the previous section (section 8.2.4) that the System is effective, that is it is capable of an effective implementation of monetary policy, including the effective realization of a uniform monetary policy stance in all parts of the euro area. There is a single money market rate and volatility is small, also thanks to the averaging facility built in the minimum reserve system.
8.3.2 Possible improvements in terms of checks and balances
Possible Treaty changes
1. We note that 12.1c (decentralization of operations) while clearly showing a bias in favour of decentralization, is in fact open-ended. In theory both extremes are possible: complete centralization and complete decentralization. It would have been possible to introduce some permanent elements in the division of labour without giving up flexibility. While too much flexibility leads to too much uncertainty (and inefficiencies resulting from that) and to NCBs and the ECB seeing each other as competitors (possibly leading to sub-optimal cooperation, because of the fear of ‘winner takes all’), too little flexibility would lead to a situation in which the System could be unable to adapt the division of labour in response to changes in the external environment. The following might introduce some more certainty. One could have laid down in the Statute that NCBs are allowed to offer standing facilities to local banks according to rules and guidelines of the Governing Council. One could also lay down that central banks are always allowed to offer payment services related to monetary policy operations to their banks. This formulation would leave open developments towards, e.g., a primary dealer system or other market led forms of organization among commercial banks (provided these banks continue to fulfil their minimum reserve obligations). At the same time, it could be clarified that the ECB may hold and manage its own foreign reserves without having to take recourse to the NCBs as an agent, or – alternatively – it should be allowed to use one or a few of the existing NCBs, in which case it could rely exclusively on this NCB (or these NCBs) for being informed of market developments on a continuous basis in the same way as the Board of Governors of the Federal Reserve System is informed by the New York Fed.
2. While the distribution of banknotes is covered by Art. 12.1c, the issuance of banknotes is not, at least this is less clear. In this respect Art. 16-ESCB is open-ended, as it allows for both extremes, i.e. that banknotes are issued only by the NCBs or only by the ECB. On top of this, while the fungibility (interchangeability) of banknotes is generally accepted and de facto a reality, a single legal basis is missing. One way of expressing that banknotes are legally fungible is to insert a text, possibly in secondary legislation, stating that the banknotes issued by any component of the System belonging to the euro area are a ‘joint and several liability’. An alternative, or possibly a combination, is to insert in the Treaty the commonly used term ‘eurosystem’, encompassing the ECB and the NCBs of the Member States without a derogation or opt-out and to print the name ‘eurosystem’ on the banknotes possibly in the form of a multilingual seal, instead of printing ‘ECB’. This would leave 16-ESCB unchanged, but it would make this article less open-ended by eliminating the possibility that the euro banknotes are only issued by either the ECB or the NCBs and it would clearly anchor the seigniorage to the System (reducing the risk that the System’s monetary income would be claimed by the Commission, which would reduce the System’s financial independence).
1. In the area of international representation (6) we have noted a possible bias in favour of the ECB. Indeed, when international fora invite the ESCB to a meeting, only sending NCB representatives would not contribute to the System’s unity. This however does not imply that in some cases the Governing Council could not decide to nominate a representative of both the Executive Board and of one of the NCBs, preferably someone with a special position covering international relations or the markets, or with special expertise. Sharing such position with NCBs will win the ECB a cooperative attitude by NCBs.
2. In a federal system committees are an efficient tool for exchanging ideas among the regional central banks and the centre for fostering convergence of views. The present committee structure of the ECB according to which all NCBs participate in each committee might lose some of its present effectiveness when the euro area is enlarged with ten or more new (accession) countries. One improvement might be to replace a number of these committees by small committees composed of a limited number of NCBs (level, e.g., vice-presidents) suppleted with an ECB member of equal ranking, thus following the example of the Federal Reserve (see appendix 2, following this cluster). Membership of these committees could rotate and should preferably be linked to expertise. Decision-making should be retained for the Governing Council. These small high-level advisory committees should discuss strategic matters and issues relating to the cohesion of the system. The advantage of small committees over the existing committees, in which all NCBs are represented, is that members of small committees are less inclined to defend ‘national’ positions and therefore are more apt to follow innovative lines of thought (bottom-up innovation). In large committees delegates will be more inclined to defend the interest of their NCB, because they expect others to do the same. Such group dynamics are less likely to occur in small committees. The aim of smaller committees is to mobilize ideas without immobilizing decision-making.
The high-level committees are an instrument for keeping the federal character alive and make best possible use of it. They could also lead to forms of specialization, as NCBs could start focussing on the development of operational tasks in only a limited number of functional areas of the System and they could become more willing to follow the example of a few leading NCBs. This is a natural way to reduce the main disadvantage (i.e. possible cost inefficiencies) of decentralized implementation, which has merits of its own. One could make an exception for committees covering areas where national authorities are also major players (e.g. international relations, supervision and oversight) or for core business like monetary strategy evaluation or amendments to the General Documentation of ESCB monetary policy instruments and procedures.
These committees could retain full composition. The other, small committees could establish, following the example of the Federal Reserve, sub-committees of varying size, when necessary including all NCBs (though the introduction of liaison officers in the high level committees should make this less necessary). An alternative would be to retain the present committees, even when growing to twenty-five members or more, while working with small-sized task forces. A disadvantage of that model would be that each proposal by a task force would still need to pass both the full-sized committee and the Governing Council, which for important non-monetary issues might want to stick to consensus decision-making, thus not solving our problem. Another aspect is that in very large sized committees the chairperson collects automatically a dominant position, while according to current practice the chairperson is usually a person from the ECB. Smaller committees allow for chairmanships to be more evenly spread over the ECB and NCBs, thus increasing the awareness that one is part of a system (a ‘team’).
Other improvements (not related to checks and balances)
Finally we want to recall we made a few suggestions for strengthening the contribution the ESCB could make to financial stability. We made suggestions for improved exchange of information between prudential and financial stability supervisors and to allow for some limited co-involvement of the ECB in the supervision of system-relevant banks (see section 8.2.1). Article 7 of the German Banking Law (Gesetz ueber das Kreditwesen) provides an example in this respect: ‘The Deutsche Bundesbank and the Banking Supervisory Office shall exchange observations and findings which may be of importance for the performance of their respective functions.’ Designating the ECB co-supervisor in a limited number of cases would make sure the ECB (in this case specifically the Executive Board) has the best possible knowledge of the positions of the largest financial players and their interlinkages. It would not imply the ECB taking over final supervisory responsibility, which would and should remain national.
We have noted that the Statute could have been phrased in such a way as to leave the division of operational functions in a less open-ended way. The Statute could be improved in this respect only through an amendment of the Treaty. Central banks might view this as opening the box of Pandora, but it is a possibility. It is not urgent as the System is functioning well, though the likely accession of up to ten new Member States to the euro area over the period 2007-2012 might make this issue more urgent later this decade. The common (shared) issuance of banknotes could be addressed as of the moment the present draft Constitution Treaty becomes effective. For the short term priority should lie with the improvements of a more practical nature. In this regards we think the spirit of co-operation is best served by introducing a workable committee structure which does not give a leadership’s role automatically to the ECB.
Furthermore, we recall our suggestion to increase the efficiency and effectiveness of the arrangements in the area of preserving financial stability by making better use of the ECB (by making the ECB co-supervisor for large banks) without taking away responsibility from national supervisors, a number of which are central banks. This could be implemented at very short notice.
Appendix 1:Developments in the Fed’s early years (1913-1935) 
The federal design of the Federal Reserve System can be explained by both technical and political factors, which are described below. We will see that in some respects the design was not perfect. This led to an initial domination by the FRBs, which was corrected by the Banking Act of 1933 (which tried to improve the coordination among the FRBs) by the Banking Act of 1935, which gave the Board a stronger grip on the conduct of open market operations (OMOs). The contents of the Banking Acts are described in a separate box at the end of this appendix.
The Federal Reserve Act of 1913 should be seen against the central-bank-less situation of these days (the charter of the Second Bank of the United States had expired and had not been renewed in 1836). The National Banking Act of 1863 had provided for the creation of nationally-chartered banks, which were effectively handed the monopoly on issuing banknotes. However, bank runs kept recurring, even in times of prosperity. And interest rates showed large seasonal swings because of the following mechanism: after the harvest season surplus funds held by banks in the agricultural areas would be sent to money centre banks in the large cities (mainly New York), where they earned money. The money centre banks often lend funds out to speculators in the stock market. When country banks needed currency, they would draw down their reserve accounts with their reserve city banks. Those banks, now with less vault cash, were compelled to draw down their own reserve accounts with their central reserve city banks, multiplying the effects. The money centre banks called in their loans and money panics could occur, which could not be assuaged by the lack of a lender of last resort. Also, national bank note issuance was inelastic, as these notes had to be secured by government bonds, bank holdings of which showed even a countercyclical pattern. When interest rates rose, existing bond holdings became less attractive and were sold, thus reducing their basis for issuing banknotes. Another problem prior to the FRA involved the collection of checks, which could take weeks.
A solution was found by allowing bank to borrow liquidity by discounting commercial bills to a Federal Reserve Bank, while the dominance of the New York financial markets was reduced by creating a number of FRBs over the country – implying the unavoidable waves in liquidity needs in the country were not amplified anymore in the New York with their spill-over effects to other parts of the country. The political advantage of creating a system of FRBs, instead of one central bank, was that it allowed for a mixture of private-public governance, with the FRBs owned by private banks (a wish from the financial community, supported by the Republicans in Congress), while the rule-setting Board had the character of a governmental, though independent, agency (a strong wish of president Wilson’s Democratic Party). We describe the features of the FRA in more detail below.
The Federal Reserve Act of 1913 provided for several major innovations: commercial bills could be discounted at the Federal Reserve Banks (thus providing for an ‘elastic currency’); member banks were required to hold their reserves at their FRB (thus ending the pyramidization of reserves in New York); a par collection system for check clearance was set up (thus ending the practice of discounting checks drawn on peripheral banks) and FRBs would be owned by its member banks. Rediscounting, or making loans to member banks, was seen as the chief means of affecting credit. The founders of the Fed had placed limits on the type of securities eligible for rediscount: eligible were discount notes, drafts and bills of exchange issued or drawn for agricultural, industrial or commercial purposes, and bank acceptances. The FRA explicitly prohibited the rediscount of notes, drafts or bills for speculative purposes, defined as notes, drafts or bills issued or drawn for the purpose of carrying or trading stocks, bonds, or other investment securities. In this way they hoped to prevent Federal Reserve credit from being used for speculative purposes. The founders expected the Fed to supply a sufficient volume of credit to accommodate growth and fluctuations in the real activity. The FRA also allowed banks to issue bankers acceptances to finance foreign trade, and it encouraged the development of a U.S. acceptance market by permitting FRBs to acquire acceptances by discount or open market purchase. The FRA also authorized FRBs to buy and sell U.S. government bonds and notes (federal, state and communal). This provision was intended to allow them a source of income in the case the income from rediscounts and the provision of services was insufficient to meet expenses. In other words, open market operations were not seen as a monetary policy instrument, but rather as creating commercial opportunities for the FRBs.
It was not until 1921, that the Reserve Banks (that had ample liquidity due to the massive gold inflow during the First World War and the 1920 downturn) began to buy and sell government securities on a significant scale. Discount window loan volume had dropped sharply in 1921-1922, when the FRBs had raised their discount rates – which had reduced their income. Only then policy-makers started to realize that the potential effect of open market operations as a monetary policy tool. The possibilities for such a policy were especially recognized by Benjamin Strong, the powerful governor of the New York FRB.
According to Wheelock (2002) the accompanying change in the composition of Federal Reserve credit (assets) during the 1920s reflects the evolution from the self-regulating. selective credit policy envisioned by the Fed’s founders toward a modern monetary policy. To take an example, the volume of discount borrowing and advances of the FRB of New York came down from dollar 871 million in 1920 to dollar 164 million in 1924. In contrast, its volume of holdings of U.S. government securities rose from dollar 47 million in 1923 to dollar 742 million in 1935.
Strong saw a need to coordinate the operations of the Reserve Banks, but he also argued they should be undertaken by the New York Fed, since the market for government securities was located in New York. In May 1922, at the request of the Federal Reserve Board, the governors of the four eastern districts (New York, Chicago, Boston and Philadelphia – later extended with Cleveland) organized among themselves a committee ‘to execute joint purchases and sales and to avoid conflicts with orders for Treasury account’. In March 1923 the Board reinserted its influence by disbanding the governors’ committee and re-establishing it as a committee under the aegis of the Board. Operating in accordance with principles and regulations set forth by, and under the general supervision of, the Board, this Open Market Investment Committee (OMIC) of the Federal Reserve System was given ‘the duty … to devise and recommend plans’ for open-market operations. The Board’s resolution establishing the OMIC provided that open-market purchases and sales should be made with the primary regard for the accommodation of commerce and business and to the effect of these purchases and sales on the general credit situation.  The wording made clear that open market operations in government securities fell under the umbrella of monetary policy making. Beginning December 1923, a System account (then called Federal Open Market Investment Account) was established, in which every FRB had a share, based on its share in the FRS’s total capital subscription by member banks. Individual banks still engaged in independent operations which the Committee executed on their behalf, but they were generally small in amount. The System account was operated by the New York Reserve Bank.
Until 1928, ‘the New York Bank was the prime mover in Federal Reserve Policy both at home and abroad, and Benjamin Strong, its governor from its inception, was the dominant figure in the Federal Reserve System’. After his resignation (and death) in 1928 there was a vacuum in leadership. ‘The Banks outside New York, seeking a larger share in the determination of open market policy, obtained the diffusion of power through the broadening of the membership of the OMIC in March 1930 to include the governors of all the Banks.’  Open market operations now depended upon a majority of twelve rather than of five governors and the twelve “came instructed by their directors” rather than ready to follow the leadership of New York as the five had been when Strong was governor. These new other Banks were, unlike New York which was accustomed to regard itself as shaping conditions in the credit market, more likely to believe that the Reserve System must adjust to other forces. They had no background of leadership and national responsibility, and ‘tended to be jealous of New York and predisposed to question what New York proposed.’ Most of the Reserve Banks were also reluctant to follow the leadership of the Board, partly because of the Board’s personnel, partly because they still thought of it primarily as a supervisory and review body. 
Early 1932 great pressure from Congressional critics had induced the System to undertake large-scale securities purchases it should have made much earlier. When the operation failed to bring immediate dramatic improvements, the System promptly relapsed into its earlier passivity. The apparent failure of monetary policy to stem the depression led in the New Deal era to the relegation of money to a minor role in effecting the course of economic events. On the other hand, action was taken to strengthen the foundations of the American financial structure. The collapse of the banking system produced a demand for remedial legislation that led to the enactment of federal deposit insurance, to changes in the powers of the Federal Reserve system (see following paragraph and box 4) and to closer regulation of banks and other financial institutions. Other measures were that banks were restricted from engaging in securities activities (Glass-Steagall Act of 1933) and prohibited from offering interest on demand deposits.
The battle of power between the New York Fed and the Board and the apparent failure of the Federal Reserve to get its act together led to changes in the FRA, incorporated in the Banking Acts of 1933 and 1935. The 1933 Banking Act eliminated the power of Banks to buy and sell government securities for their own account except with the explicit permission or at the direction of the Federal Open Market Committee (FOMC), which replaced the Open Market Policy Conference, and consisted of the twelve heads of the Banks. The Banking Act of 1935 further centralized the authority and responsibility for open market policy by changing the composition of the FOMC: it would be composed of the seven members of the Board and five representatives from the FRBs, each to be selected annually by the board of directors of stipulated groups of Reserve Banks. New York has a permanent vote, the other eleven FRBs share four votes. The Banking Act of 1935 did more: it changed the name of the Federal Reserve Board to the ‘Board of Governors of the Federal Reserve System’;  reconstituted the Board by eliminating ex officio members; and strengthened the control of the FOMC over open market operations.
Box 4: Banking Acts of 1933 and 1935
These acts are of special relevance for the institutional design of the Federal Reserve.
Banking Act of 1933 
– Establishment of the FOMC (composed of one representative of each Reserve Bank), placing the open market operations under the control of the FOMC.
– Introduction of a provision forbidding Reserve Banks from engaging in negotiations with foreign banks except as authorized by the Board.
– Introduction of a provision stating that the Board’s funds ‘shall not be construed to be Government funds or appropriated moneys.
– The Banking Act of 1933 also provided for the establishment of the Federal Deposit Insurance Corporation (FDIC) to protect small depositors against loss due to bank failure, the obligation for member banks to become member of the FDIC and a prohibition for commercial banks to engage in securities activities.
Banking Act of 1935 
– Reorganization of the Federal Reserve Board: the name was changed into the Board of Governors of the Federal Reserve System. Ex officio membership (of the Secretary of the Treasury and the Comptroller of the Currency) was scrapped, implying the U.S. President would appoint henceforth not five, but seven members of the Board. Their tenure was increased from ten to fourteen years (thus restoring the rule that the president appoints one new member only every two years, which rule had been established at the start of the Fed in order to prevent the president from being able to appoint too many new members within one presidential term of four year.)
– Chief Executive Officers at the Federal Reserve Banks: the name ‘governor’ had, until then, been reserved for the person was designated by the President from among the five presidential appointees to the Federal Reserve Board to be the Board’s active executive officer.  The chief executive officers of the FRBs had started to use the title of ‘governor’ as well, a title generally assigned to the operative head of central banks. The Banking Act of 1935 stipulated that the chief executive officers of the Federal Reserve Banks should henceforth be called presidents, the title ‘governor’ being reserved for each member of the Board of Governors. The appointment of the presidents by the board of directors of their FRB would henceforth be subject to approval by the Board of Governors.  They would be appointed for five-year terms. Before, their term had not been fixed and they could have been dismissed ‘at pleasure’ by their boards. The Board of Governors would be chaired by the ‘Chairman’, designated from among the governors by the President of the United States, by and with the consent of the Senate.
– New Federal Open Market Committee: Eccles wished to take away the power over open-market operations from the privately run FRBs, acting through their governors. He wished the power to be concentrated in an Open Market Committee of the Federal Reserve Board in Washington D.C. In the House version of the bill the Open Market Committee was formed of Board members alone. The governors of the FRBs (renamed presidents – see previous indent) would annually chose five of their number to act in an advisory capacity to the committee. In the final version they (i.e. five FRB presidents) had become full voting members of the committee. Eccles still valued the outcome because it established the principle that open market operations would be initiated in Washington. Also no FRB would be permitted to engage, nor decline to engage, in open-market operations without approval by the committee. Eccles also valued that ‘records were to be taken and kept of each vote on open-market policy questions, and with a full account of such actions being submitted annually to Congress, there could no longer be any buck-passing.’ 
– The Board’s powers to alter reserve requirements were enlarged. It was given the authority to change member banks’ reserve requirements between the minimum percentages specified in the act of June 1917 and twice those percentages. Thus the Board of Governors was given authority to use the reserve requirements as a monetary instrument without approval by the President.
 See e.g. Memorandum-Balladur (29 December 1987), in which Balladur mentions the possibility of a single currency and ‘a common central bank and ‘Bundes’banken in each country.’ (HWWA (1993), p. 338.)
 The argument that financial stability is an issue covering more than one geographical area (e.g. the European Union or the G10 countries), while true, should not be an argument not to improve regional cooperation.
 We do not see advantages of the ECB as sole supervisor, or as co-supervisor of purely local banks, as in those cases the existing information asymmetry (i.e. the fact that local supervisors have better access to relevant local information) is not outweighed by other benefits, like increasing returns of scale or information of system-wide relevance.
 This suggestion relates to regular information sharing, which contributes to crisis prevention. In times of an actual crisis supervisory authorities and NCBs should be (and in fact are) authorized to share confidential information relevant for crisis resolution.
 An effective functioning of the ECB in this respect also requires that NCBs inform the ECB on a continuous basis or any unusual large access to their local marginal financing facility, other organized lender-of-last-resort assistance and large interbank (including cross-border) exposures – see Van den Berg and van Oorschot (2000).
 See also Art. 10.2-ESCB in cluster III.
 This last idea was captured in the second paragraph of Art. 12.1 and in Art. 14.3 of the ESCB Statute.
 Compare the perceived influence of Wall Street in the United States and of the ‘City’ in Britain.
 In some cases this has led to complicated institutional–legal frameworks, e.g. in the case of the Irish central bank, because one has to bring Treaty-based functions and attributed (or delegated) national functions under one governance structure.
 Art. 27 and 28 have been dealt with in cluster I because of their relevance for the System’s financial independence vis-à-vis the outside world. Art. 29 is less relevant for cluster II; it is a technical article determining each NCB’s share in the ECB’s capital. It is thus more relevant for the relation between NCBs than between the NCBs on the one hand and the ECB on the other.
 In the US the FRA authorizes the Board of Governors of the Federal Reserve to prepare policy positions.
 The relation between the governors and the Executive Board will be dealt with in cluster III.
 The genesis of the Delors Report and Art. 30-ESCB show the transfer to the ECB was meant to create a visible centre with a significant amount of foreign reserves to ensure the credibility of the System’s exchange rate policy.
 Strictly speaking this depends on the remuneration of the concomitant claim of the ECB on the NCBs.
 Reverse transactions are executed through (standard or quick) tenders. Tenders are announced by the ECB, bids are collected through the NCBs, allocations are decided by the ECB. Outright transactions are executed through bilateral procedures. See ECB General Documentation on ESCB monetary policy instruments and procedures.
 Only in exceptional circumstances (as defined by Governing Council).
 Only in exceptional circumstances (as defined by Governing Council).
 Each NCB manages a portion of the foreign reserves of the ECB corresponding to its share in the transfer of reserves to the ECB, on the basis of an agency construction. The ECB (Governing Council) issues detailed guidelines (including benchmarks) for the management of these reserves. Size end 2002: euro 44.8 billion (excluding euro 7.8 billion gold).
 The ECB contacts the counterparties in the financial markets. Foreign currency is moved to/from account liquid account of the ECB held at the NY Fed. The euro leg of the transaction is settled through the books of an NCB. (This gives rise to intra-eurosystem claims/liabilities in the books of the ECB and the NCB involved.)
 The ECB manages a euro securities portfolio, the size of which corresponds more or less to the size of its capital and reserves (euro 4.4 billion end 2001).
 Only for a few international organizations.
 The ECB has an advantage, because for instance in the context of G7 meetings the ECB’s president or an Executive Board is invited to be present, and not an NCB president (though the G7 governors might be present in their own capacity).
 ECB has observer status at the IMF. The IMF is an intergovernmental organization, of which only countries can be a member. The ECB is entitled to hold IMF reserve positions and SDRs (Art. 30.5)
 Unlike the NCBs, the ECB has an observership at the IMF. NCBs are usually represented through secondment of staff to their country’s or constituency’s Executive Director’s office.
 Of the consolidated eurosystem balance sheet of ultimo 2001 claims on credit institutions arising from monetary policy operations constitute 25%, total forex holdings (including gold) 51% (of which 7 percentage point is held by the ECB), and NCB held domestic euro denominated assets 11%, on a total balance of euro 814 billion (ECB Annual Report 2001, p. 200).
 European NCBs are not involved in check collection, like their American counterparts.
 On the basis of a 1994 Board of Governors decision the chairman is the ex-officio director on behalf of the FRS and the president of the New York Fed is the appointed second American member of the BIS Board. The American shares in the BIS were owned by commercial banks.
 This and the following quotes are taken from BIS (1963), Eight European Central Banks, p. 60ff. This division of labour still applies to the present Bundesbank, though the number of LZBs has become smaller and monetary policy is formulated by the ECB’s Governing Council.
 The Landeszentralbanken are branches of the Bundesbank. Although they do not have there own balance sheet, they are allocated specific tasks according to the Bundesbank Law. Despite losing their legal independence, when the Bank deutscher Länder was being replaced by the Bundesbank, the LZBs retained their organizational independence. The LZBs conduct on their own responsibility all transactions and administrative business within the area for which they are competent. See also appendix on the Bank deutscher Länder in cluster III.
 The LZBs collect the bids of small banks for participation in the refinancing operations of the Bundesbank sometimes even through their branch offices.
 See footnote with Art. 12.1c, section I.2.
 In a primary dealer system there would be less need for NCBs feeding the centre with projections of local factors affecting the euro area’s overall liquidity system. In such a world the centre would probably rely more on information contained in market prices and forward rates, assuming the markets would make sure they are well-informed about important local tax pay days and other technical factors. A difference with the Fed would be that the eurosystem relies to a larger extent on the buffer function of the minimum reserve accounts and mostly relies on weekly (and not daily) refinancing operations.
 See appendix 2 for specialization within the Federal Reserve System.
 Amtenbrink (1999), p. 94.
 See box 4 in appendix 1.
 Friedman & Schwartz (1963), p. 445).
 Eccles (1951), p. 167-174.
 M. Goodfriend (2000).
 Art. 3.3 indirectly protects NCBs with supervisory responsibilities, because it stops short giving the ECB responsibilities of its own in the area of prudential supervision.
 Art. 32 stipulates that the monetary income flows to the NCBs (and not to the ECB).
 Art. 33 determines that the ECB’s surplus profits flow to the NCBs. Losses of the ECB can be covered by NCBs, but only when the Governors so decide (weighted voting). The ECB may build reserves, but it is not free to spend, its budget requiring approval by the Governing Council.
 There is a link with cluster III, because the actual division of labour is laid down in the General Documentation (GD), which is approved by the Governing Council, the composition of which is determined in Art. 10.1-ESCB (see cluster III). The GD is prepared by ESCB committees, comprising both ECB and NCB representatives.
 Art. 16 has been used to give the ECB more ‘own’ funds, though the NCBs remain the ECB’s shareholders and the ECB’s budget needs to be approved by the Governing Council.
 The patrimonial decisions (in which the votes of the Executive Board members have zero weight – see Art. 10.3-ESCB in cluster III) do not count as such as they are aimed at decisions which affect the NCBs’ relative financial positions, though there can be side-effects for the ECB’s income position (esp. Art. 30.3).
 Art. 26-ESCB and Art. 15 of Rules of Procedure respectively – see under Art. 12.3-ESCB for Rules of Procedure. (External auditors examine the books of the ECB – see Art. 27-ESCB.)
 Apart from making sense for practical reasons, this division of labour is the product of the efforts by some Member States to give the ECB a visible role in the area of exchange rate policy, while other Member States wanted to prevent any remaining national influences over exchange rate policies.
 We repeat that a continued operational role for the NCBs is also essential for the input of their governors in the Governing Council.
 In the meantime in the draft Constitutional Treaty, as signed by the Heads of State and Government on 29 October 2004, but still to be ratified, the term ‘eurosystem’ has been defined in Art. 1 of the ESCB-Statute and will have formal Treaty-status, once the Constitution becomes effective.
 See chapter 8.2.1 (Art. 3.3-ESCB) and sections I.1 and I.2 of Art. 3.3-ESCB.
 This appendix is based for the most part on Friedman & Schwartz (1963), A Monetary History of the United States, 1867-1960, p. 251-253; 362-391; 411-420; 445-449; Wheelock (2002), ‘Conducting Monetary Policy Without Government Debt: The Fed’s Early Years’, in: The Federal Reserve Bank of St. Louis Review, May/June 2002, p. 1-14; Federal Reserve Bank of Chicago (1988), Annual Report 1988, ‘Historical Perspectives: The Bank at 75’. For a discussion of the checks and balances involved see chapter 6 above (section on the Federal Reserve).
 See Art. 16-ESCB, section I.2.
 Moore (1990), The Federal Reserve System – A History of the First 75 years, p. 4-5; FRB of Boston (1990), p. 13-15.
 FRA (1913), Section 13 (as amended in 1916 for the purpose of including a wider array of bankers acceptances).
 FRA (1913), Section 13. An exception was made for bills drawn for the purpose of trading bonds and notes of the Government of the United States.
 This thinking reflects the so-called the Real Bills Doctrine. Friedman is critical of the Real Bill Doctrine. It did not work symmetrical, because commercial paper was abundant (so currency did not necessarily move down with the volume of commercial paper). Furthermore, in practice the doctrine was not pure, because not only commercial paper was accepted, but indirectly also government paper (because government paper was used as collateral for 15-day notes issued by banks, which itself was accepted as commercial paper. Finally the doctrine overlooked the need to take into account other sources of money creation. We quote F&S on this (Friedman & Schwartz (1963), p. 191-193): ‘The road to elasticity of the money stock, it was believed, was paved with commercial paper that member banks would present for rediscounting. [….] Conflicting monetary policies were sometimes pursued, because it was not clear to the Federal Reserve that the rediscounting of any security, open market purchases, and gold inflows, all had precisely the same effects on the money stock as the rediscounting of eligible paper.’ The risk of bank runs though was effectively reduced, because banks to generate cash quickly in times of need by discounting paper at an FRB. A further reflection of the Real Bills doctrine was that the FRBs had to set discount rates ‘with a view to accommodating commerce and business’ (FRA, Section 14(d)). For the Real Bills doctrine, see also section I.2 of Art. 12.1c-ESCB.
 Wheelock (2002), p. 5. Section 14 of the FRA (1913) allowed open market transactions in securities eligible for rediscount, gold and all kinds of government paper.
 Strong directed two major monetary policy operations during the 1920s involving substantial open market purchases in 1924 and 1927. Scholars disagree whether his primary motivation had been to ease money market conditions or to redirect the international gold flow away from the United States toward the United Kingdom, in an effort to help Britain restore, and then preserve, gold convertibility of the pound. His initiative irritated members of the Federal Reserve Board, who believed the committee of governors (see below) had overstepped its authority. They also believed that credit should be provided to banks for selective purposes, i.e. through discounting commercial and agricultural loans. Otherwise, the Fed risked contributing to speculative attacks. The stock market was roaring, but most Board members were reluctant to constrain this by raising the discount rate, because this would not only punish investment brokers, but also businesses. However, the FRBs believed it neither practical nor desirable for the Fed to affect the private allocation of credit. In 1928 the Federal Reserve’s open market committee decided to implement a more restrictive monetary policy. The FRBs also began to increase discount rates – and for the most part these actions were supported by the Board. Despite this, credit continued to flow to the stock market. Some officials recommended that banks be required to liquidate speculative loans before being permitted to rediscount eligible paper. When the stock market declined sharply in October 1929 and after, the Fed did not respond aggressively, one reason being that some FRB governors were of the view that open market operations constituted artificial easing and would lead to a dangerous misallocation of credit. They were not able to prevent open market operations altogether, but their attitude slowed the Fed’s response to the Great Depression. (Wheelock (2002), p. 6-11.)
 Wheelock (2002), p. 6.
 Federal Reserve Bank of New York, Annual Reports over 1920, 1923 and 1935.
 Prochnov (1960), p. 119.
 See Dykes and Whitehouse, ‘The Establishment and Evolution of the Federal Reserve Board: 1913- 1923’, in: Federal Reserve Bulletin, April 1989, p. 24.
 FRA (1988), Section 12A(c). See also Marriner Eccles (1951), Beckoning Frontiers, p. 169.
 Dykes and Whitehouse (1989), p. 241. The wording seems to have been borrowed partly from the last part of Section 14(d)-FRA (1913): “[Every Federal reserve bank shall have the power:] (d) To establish from time to time, subject to review and determination by the Federal Reserve Board, rates of discount to be charged by the Federal reserve bank for each class of paper, which shall be fixed with a view of accommodating commerce and business.”
 Later called System Open Market Account (SOMA).
 Friedman & Schwartz (1963), p.251n. They also mention that Strong and the other Bank governors regarded the coordination of open market operations as a voluntary agreement. Reserve Banks could refuse to participate in operations recommended by the committee. The issue was never pressed to a final decision and became irrelevant after the Banking Act of 1935, which would lead to new decision-making rules in the FOMC, the committee which had been established by the Banking Act of 1933. The voluntary character of the arrangements had been criticized by Eccles who, when asked by president Roosevelt to become the chairman of the Board, had produced a critical memorandum which would lead to the Banking Act of 1935 (see Eccles (1951), p. 165-176). Another issue never pressed to its end was how much leeway an FRB had in maintaining an unchanged discount rate, while the rest of the FRBs with the approval of the Board wanted to change their rates. In 1927 Strong was leading an effort to reduce the rates, basically for international reasons. Chicago resisted and was ordered by the Board to reduce its rate. Chicago complied, but announced at the same time it would seek an opinion from the U.S. Attorney General, as to the legality of the Board’s action. In the end Chicago did not pursue this legal course. FRB of Chicago, Annual Report 1988, p. 18.
 See also Prochnov (1960), p. 121. The OMIC was redesignated as the Open Market Policy Conference.
 See Friedman & Schwartz (1963), p. 414-416. Strong’s successor at New York (Harrison) did advocate a strong reaction to the unfolding liquidity crisis, but he failed to convince the others.
 Friedman & Schwartz (1963), p. 419.
 FRB of Chicago, Annual Report over 1988, p. 20.
 Friedman & Schwartz (1963), p. 446n. Banks were still allowed to purchases government securities for their own account, subject to certain restrictions, in an emergency involving individual banking institutions. This possibility would be scrapped by the Banking Act of 1935.
 Prochnov (1960), p. 121.
 Formally only since 1942, see footnote 39 below and Art. 10.2, section I.2 in cluster III.
 Before, only the executive head of the Board also had been a governor and addressed as such. The other members were simply members of the Board and were addressed without title. The chief executive officers of the Banks had been governors too, but henceforth became ‘president’ of their Banks. It was through this new name (Board of Governors of the Federal Reserve System) that the term ‘Federal Reserve System’ appeared in the FRA. The system as such was not a legal entity and the FRA had described the components of the system, their powers and the way they relate. In fact, the recital of the FRA does not refer to the establishment of a system, but to the establishment of Federal reserve banks: “An Act. To provide for the establishment of Federal reserve banks, to furnish an elastic currency, to afford means of rediscounting commercial paper, to establish a more effective supervision of banking in the United States, and for other purposes.” See also section I.2 under Article 1-ESCB of Cluster I.
 Another famous act from this era is the McFadden Act of 1927 which prohibited interstate banking. The Riegle-Neal Act of 1994 again permitted (adequately capitalized and managed) bank holding companies to acquire banks in any state.
 The Banking Act of 1933 is sometimes referred to as the Glass-Steagall Act of 1933. In practice, the Glass-Steagall Act of 1933 has come to mean only those sections of the Banking Act of 1933 that refer to banks’ securities operations – sections 16, 20, 21 and 32 -, which establish a separation between the banking and securities businesses. The Glass-Steagall Act was de facto repealed when in 1999 Gramm-Leach-Bliley Act was passed. The GLB Act also allowed affiliation between banks and insurance underwriters, thus reversing part of the Bank Holding Act of 1956 The GLB Act allows banks (financial holding companies) to offer a menu of financial services, including investment banking and insurance sales..
 See FRA (1988), Section 14(g). According to Kettl (1986), p. 46, this was an undisguised slap at the extensive negotiations Strong had with Norman, the governor of the Bank of England. Though German Länder may, in their field of competence, conclude public international law agreements with third countries, subject to approval by the Federal Government, this does not apply to the Landeszentralbanken. (Zilioli and Selmayer (1999a), p. 278.)
 In the wake of the Depression, Congress wanted to strengthen the Fed, and the committee report on the bill noted Congress’s intent to allow the Fed to set its own management policies. This clause ended GAO’s audit of the Fed. (Kettl (1986), p. 154.) In 1978 the GAO audit would be restored, be it with a limited mandate – see Article 27-ESCB.
 This content of this part of the box follows closely Eccles (1951), p. 222 – 229. Eccles was chairman of the Federal Reserve from 1934 to 1948. In September 1934 Eccles was sounded by president Roosevelt about the post of chairman (Governor) of the Federal Reserve Board. Eccles replied that the post would be an appealing one only if fundamental changes were made in the Federal Reserve system. He elaborated his ideas in a memorandum, presented to and discussed with president Roosevelt on 4 November 1934. Roosevelt supported his ideas. Subsequently Roosevelt set up an Interdepartmental Banking Committee, consisting of the heads of agencies involved in banking operations and chaired by Treasury Secretary Morgenthau (Eccles (1951), p. 165-176 and p. 194).
 According to Eccles (1951), p. 216 and 222, it was at the insistence of Senator Glass, himself a former Secretary of the Treasury, that the ex officio membership of Treasury Secretaries was dropped, as Glass knew by experience he had had too much influence on the Board’s decisions. Treasury Secretary Morgenthau then insisted that the other ex officio membership, i.e. that of the Comptroller of the Currency (a subordinate in the Treasury Department), should also be dropped.
 FRA (1988), Section 10(1) and 10(2).
 The first group of governors would be appointed for periods from two to fourteen years, so that not more than one would expire in any two-year period.
 The ‘governor’ of the Federal Reserve Board did not act as chairman. Section 10 of the FRA of 1913 stipulated that ‘[t]he Secretary of the Treasury shall be ex officio chairman of the Federal Reserve Board.’
 The same applies to the first vice president.
 Eccles had promoted the idea of making the appointment of the FRB’s governor subject to approval by the Board in Washington, because in the day-to-day operations the governor, being the FRB’s ceo, was more important than the FRB’s chairman (one of the class C directors). According to Eccles the ceo was ‘the creation of the private interests which supplied a majority of the board of directors and who hired and fired the governor without let or hindrance of the Reserve Board in Washington.’ The importance of the governors of the FRBs had increased with the increased importance of the open market operations, which was coordinated by a committee consisting of governors. (Eccles (1951), p. 168/9.)
 FRA (1913), Section 4. FRA (1988), Section 4, seems to apply only to ‘other’ officers and employees.
 The same procedure applies to the designation of the Vice Chairman.
 Eccles (1951), p. 174.
 Section 205 of the Banking Act of 1935 stipulated that the five representatives (together with one alternate for each) are to be selected annually by the boards of directors of the FRBs according to the following schedule: ‘one by Boston and New York; one by Philadelphia and Cleveland; one by Richmond, Atlanta and Dallas; one by Chicago and St. Louis; and one by Minneapolis, Kansas City, and San Francisco.’ This was changed in 1942 (amendment to Section 12A FRA, 7 July 1942, 56 Stat 647) defining the following groups: ‘(1) New York, (2) Boston, Philadelphia, and Richmond, (3) Cleveland and Chicago, (4) Atlanta, Dallas, and St. Louis, (5) Minneapolis, Kansas City, and San Francisco. According to the Annual Report of the Board of Governors of the FRS of 1942 (p. 56) ‘[t]his arrangement makes provision for continuous representation of the FRB of New York on the FOMC. This is for the reason that the New York Bank is in the principal capital market and acts as the agent for the FOMC in the operation of the System open-market account.’ Before 1942, New York had always had a vote, implying Boston never had one. In 1942 Boston was shifted to one of the other groups, while at the same time some reshuffling took place, each group still consisting of geographically neighbouring districts. See also Article 10.2-ESCB in Cluster III. Most publications of the Federal Reserve contain a map showing the districts.
 According to Friedman & Schwartz, the change in name of the Federal Reserve Board into the Board of Governors was the final seal on the transfer of effective power from the Banks to the Board. (Friedman & Schwartz (1963), p. 445-447.)
 Eccles (1951), p. 224/6. FRA (1988), Section 10(b); see also Art. 10.4-ESCB, section I.2.
 Since 1933 the Board had been able to increase or decrease reserve balances member banks were required to maintain against their demand and time deposits, during special circumstances and only with the President’s approval. Before 1933 the Board had only been allowed to suspend reserve ratios for limited periods. It could not change them. This would only have been possible through amending the FRA, Section 19. See Friedman & Schwartz (1963), p. 447, also for some later amendments of Article 19-FRA.
 Eccles (1951), p. 226.