How centralised should the operation of a central bank be? Central banks were originally created as instruments to facilitate the financial arrangements of unified and centrally directed states, as was the case for the first central banks – in Sweden, England, and France.
On the other hand, models for federal central banking came rather later, in Germany (1875), Switzerland (1907) and the US (1914). The federal central banks required complex rules to ensure that there was no influence over its direction by the federal government, and that operations in regard to monetary (discount) policy and transfers between different parts reflected the diverse conditions of a large geographic area.
The central banking side of the US federal model, the Federal Reserve System, has often been held up as a model for the European System of Central Banks.
The Federal Reserve System
The question of the relationship of a central federal bank to local banking systems – and to the patronage systems built up by local elites – has always been a highly contentious issue in US politics. The feeling that local interests would be sacrificed to a Massachusetts and New York banking elite was a strong driver of opposition to Alexander Hamilton’s plans of 1790.
The Federal Reserve System was designed to preserve checks and balances, and to ensure that the system could be neither dominated by the powerful east-coast financial community nor by the federal government in Washington.
The regional Federal Reserve Banks corresponded to what were felt to be logical economic areas, not necessarily overlapping with state boundaries. A separate reserve bank for each state would have created an over-complicated system, with a large and unwieldy central committee (originally termed the Federal Reserve Advisory Council). The majority on the boards of the reserve banks were selected by the local nationally chartered banks, which comprised the US financial system and which were required to subscribe to the capital of the reserve bank. This principle continues to the present day. The seven-member board in Washington was the political counterpart, and five members were appointed by the president with the advice and consent of the Senate.
Lessons for Europe?
To highlight the surprising character of this set up, a mental experiment might be helpful. A modern European equivalent to the Federal Reserve would be to create private sector-based regional central banks, for instance with Alpine, Baltic, North Sea, Atlantic, Danubian, and Mediterranean banks.
The original (1914) Federal Reserve System closely resembles the interaction of national central banks in the international system of the gold standard in many ways. The 12 reserve banks controlled their own operations, and had their own discount policy.
The individual banks were also required to hold gold in order to allow for the clearing of debit balances. The loss of gold would affect their reserve ratio, meaning that they would also need to reduce credit to banks, and would thus shrink the regional money stock.
In 1920-21, in the severe deflation at the end of the first world war, agricultural areas were affected more strongly than the industrial districts, and payments to farmers were slow and at low prices. The consequence was a balance of payments deficit. As the reserves fell, the district reserve banks were under pressure, but they borrowed from other reserve banks with large surpluses so as to minimise the impact.
In the Great Depression, when a similar effect might have been expected to operate, the district shortfalls as a result of regional balance of payments deficits were made good, not by interdistrict accommodation, but by federal fiscal transfers made through the Federal Reserve System.
The Federal Reserve System in practice during the Great Depression also moved away from its previous practice of limiting loans to credit secured by commercial bills (the so-called real bills doctrine) to operating much more with government securities as collateral, and subsequently to the direct purchase of government securities.
The expansion of the federal budget avoided the need for big financing operations by the central bank through the interdistrict settlement account, and the alteration in the credit practice of the system removed monetary policy from being driven by regional imbalances. Large interdistrict surpluses and deficits only appeared again after 2008, in the aftermath of the failure of the private interbank market. Then, as in Europe, the Federal Reserve System substituted for a failure of private-sector bank intermediation.
The gold-standard era approach to discount rates, in which individual member banks set their own rate, looks very different from the modern practice of monetary union, based on a single uniform interest rate.
The current crisis
For Europe, the one-size-fits-all approach meant that in the 2000s interest rates in southern European countries were too low, and in northern Europe they were too high. Identical nominal rates with divergent real rates produced unsustainable credit booms in the south. By contrast, a gold-standard rule would have produced higher rates for the southern European borrowers, which would have attracted funds to where capital might be productively used, and at the same time acted as a deterrent against purely speculative capital flows.
A modern equivalent to the gold standard/early Federal Reserve approach would require differing (higher) levels of collateral requirement for central banks operating in countries with a housing and credit boom (pre-2007 Spain or Ireland) than in countries with no credit boom (pre-2007 Germany) (Brunnermeier 2012).
The mechanism of settlement changed in the 1930s, and was renamed from the Gold Settlement Account to the Interdistrict Settlement Account. From the 1970s to 2008, the balances were of small size and limited importance, because interdistrict transfers occurred primarily through the interbank market. After 2008, with the seize-up of the interbank market, the Interdistrict Settlement Account became very significant.
Figure 1. ISA balances of US District Feds
Source: Federal Reserve Bank of St Louis.
In the aftermath of the financial crisis, large and persistent imbalances appeared, however, with the large liabilities of the San Francisco and Richmond banks, and the large asset balances of New York. They are relatively small figures compared with the overall expansion of the Federal Reserve System’s balance sheet, but they are not insignificant.
The Interdistrict Settlement Account balances are comparable to European TARGET positions in that they arise from very large movements of funds out of some commercial banks that operate across the whole of the US, but have their headquarters (and thus their financial location) in a particular place within one of the twelve Federal Reserve districts. The most plausible explanation involves the head office location of large banks in the San Francisco district (Wells Fargo) and in the Richmond district (Bank of America), with the Federal Reserve Bank keeping claims against these banks, rather than selling them in the settlement process.
Figure 2. Gross TARGET and ISA balances relative to GDP
Source: Sinn (2012a), updated.
Since the Interdistrict Settlement Account imbalances reflect fundamentally changing market perceptions of US private financial institutions, rather than current-account imbalances between the various regions of the US, and as local District Feds being private institutions are not vulnerable to local political pressure aiming at state finance with the printing press, they do not display the permanence that has characterised their European equivalents, where banks in deficit countries are paralysed because of the ties between banks and sovereigns (with banks holding the paper of the sovereigns that bail them out) (Garber 2010).
Differences between the US and Europe
The most pronounced difference between the US and the European system is, however, that only the US has a settlement system that requires the debtor District Feds to securitise their Interdistrict Settlement Account debt, i.e. to redeem their liabilities with interest bearing, marketable assets (Sinn and Wollmershäuser 2012). In the Eurozone, by contrast, the debt is simply kept in the books and carried forward year by year with interest being added. The figure above shows that during the crisis, around the month of settlement (April), the US balances normally go down significantly. An exception was 2011. In that year, settlement was postponed by a year to give the deficit District Feds more time to react.
In addition to generating useful incentives to keep the interdistrict imbalances small, the settlement may be thought of as protecting local central banks more effectively against the break-up of the system. European governments that own the central banks are vulnerable to political pressure to participate in further bailout activities such as government bond purchases and the establishment of intergovernmental rescue programs, which both reduce the TARGET imbalances.
The Federal Reserve System as a whole has a sovereign as a counterpart, while the European System of Central Banks does not. The mechanism of the TARGET claims inevitably creates a demand to move further on the path to fiscal federalism.
Brunnermeier, Markus (2012) “Macroprudential Regulation: Optimizing the Currency Area”, in The Great Financial Crisis: Lessons for Financial Stability and Monetary Policy, European Central Bank, Frankfurt.
Garber, Peter (2010), "The Mechanics of Intra Euro Capital Flight", Deutsche Bank Economics Special Report, December 10.
Sinn, Hans-Werner and Wollmershäuser, T (2011), “Target Loans, Current Account Balances and Capital Flows: The ECB’s Rescue Facility”, CESifo Working Paper 3500, June 2011, NBER Working Paper 17626, November 2011, International Tax and Public Finance 19, May 2012, p. 468-508.
Sinn, Hans Werner (2012) Die Target-Falle, Munich: Hanser.