8 - Cooperation and the Fed Swap Network

Summary

The Fed was keen the cooperate with the Bank from early in the 1960s. The Bank on the other hand took time to warm up and share information with colleagues at the Fed. But progressively swaps became a key feature of the United Kingdom's exchange management strategy. They offered cheap and discreet short-term dollar loans. The Bank started to use and abuse these loans to manage sterling.

The Gold Pool was the first palliative measure put in place after the introduction of convertibility. The second was the US swap network. Swaps refer to one central bank exchanging domestic currency against foreign currency with another central bank. For example, the Bank of England could exchange sterling against dollars with the Fed. The two central banks would decide to reverse the transaction at an agreed forward exchange rate. The goal of these swaps was to provide foreign currency when needed. The main swap partner for the Bank of England during the Bretton Woods period was the Federal Reserve. Bordo et al. estimated that almost 57 per cent of all the $15.3 billion swap contracts of the Fed between 1962 and 1971 were with the Bank of England.¹ The Bank was the Federal Reserve’s biggest customer when it came to swaps. The benefits of swaps were mutual, as Capie stressed: ‘The essence of the swap network that revolved in the main around the Federal Reserve was to provide, in the case of the Bank, dollars for intervention purposes and the Fed with sterling with which it could purchase dollars that otherwise might be converted into gold.’²

The Federal Reserve international swap network was initially designed to ease the pressure on US gold reserves. Reserves were under pressure because of the US commitment within the Bretton Woods Agreement. Swaps ended up being a tool for managing sterling. The Bank agreed to join the network as a favour to the Federal Reserve. Within a few years, swaps had become the centrepiece of British foreign exchange management. They were part of a complex network of loans put in place to help maintain the fixed exchange rate of the pound. This loan network was the joint effort of the United States and the United Kingdom. The two countries often ended up negotiating further loan agreements with the rest of Europe. Robert McCauley and Catherine Schenk have recently shown that the swap network in the 1960s was not solely used for exchange rate management.³ They also helped manage Eurodollar funding liquidity and Libor yields.

Why was the United States so keen to grant the United Kingdom access to loans? US policymakers recognised that sterling was the dollar’s first line of defence. A sterling crash would trigger a run on the dollar. It would reveal the weakness of sterling as an international currency. As a consequence, investors would in turn fear for the stability of the other international currency, the dollar. Under normal circumstances, it would be reasonable to expect that a depreciation of sterling would lead to an appreciation of the dollar. However, this mechanism did not apply. In the context of the Bretton Woods system, the two currencies had a similar function. Leadership for the main international reserve currency was moving from the pound to the dollar. It was clear to investors that if sterling fell, the dollar would come under pressure. The United States wanted to avoid a run on the dollar at all costs.

The Bank was in a position where it needed help, and this led to a warmer relationship with the Federal Reserve, unlike in the early 1950s. The Bank and the Fed were in contact by telephone every day. I present new records of these conversations to trace the evolving relationship. The swap network enhanced the stability of the international monetary system, but only in the short term. The long-run liquidity problem raised by the Triffin dilemma remained. The dollar and sterling supporting each other made the currencies inherently prone to instability. Both currencies were liable to speculative attacks. This mutual dependency made the Bretton Woods system more unstable.

The Bank of England’s Use of Swaps

The 1960s marked the rediscovery of swaps. Central bankers in Europe had used them in the 1920s.⁴ Swaps re-emerged when the Swiss National Bank suggested that the Federal Reserve establish a swap line because the Swiss National Bank was experiencing large dollar inflows. The Federal Reserve did not want these dollars converted into gold. After this initial try, the practice was broadened. The Fed established a swap network with most European central banks to act as a buffer for its gold reserves.⁵ The Fed wanted to keep these banks from using the gold window to convert their dollars into gold. The Fed could use the swap line to borrow in a European currency and buy dollars back from a European central bank that had accumulated them. The United States would also put political pressure on European countries to hold a larger proportion of their reserves in dollars. The aim was to avoid too heavy a gold drain from the United States. A gold drain would have undermined the credibility of the dollar. From 1962 onwards, swap lines were key to the US policy to defend gold and provided dollar liquidity to European central banks.⁶ They were rediscovered by central bankers in the wake of the 2008 global financial crisis.

The Federal Reserve took the initiative in creating its first swap line with the Bank of England. At the time, the Bank only agreed as a gesture of good faith. It did not see any advantage in taking part in the network. Soon, swap contracts would become the pound’s lifeline. Since the fate of the pound and the dollar were interlinked, the Federal Reserve provided this support with the aim of keeping a crisis from contaminating the dollar. The Bank of England first agreed to a $50 million swap line with the Fed, which was later increased to $500 million. A note from the governor in March 1962 about a discussion with Sir Denis Rickett from the British Treasury reads: ‘Sir Denis agreed that there was no merit in this but that it might be necessary to go along with the American proposal as a symbol of international co-operation.’⁷

Despite early reluctance for their use, swap agreements became the centrepiece of British exchange rate policy in the 1960s. The swap line started with a small drawing by the Federal Reserve on the $/£ swap line in 1962–63. After that, drawings were only made by the Bank of England to acquire dollars until 1971. Table 8.1 presents annual drawings by the Bank. It highlights the relative importance of the United Kingdom when compared with the other fourteen participants in the network.⁸ From 1964 to 1967, the United Kingdom constituted on average 80 per cent of the Federal Reserve dollar lending of the fifteen nations able to draw from the Federal Reserve loan facility.

Table 8.1. Bank of England drawing on US swap networks

In million dollars	BoE drawing % of all countries drawing	BoE drawing	All countries drawing
1962	0%	0	0
1963	0%	0	50
1964	88%	1,370	1,550
1965	100%	1,765	1,765
1966	69%	625	910
1967	66%	1,650	2,487
1968	58%	2,045	3,503
1969	46%	795	1,719
1970	22%	400	1,834
1971	0%	0	30
1972	0%	0	19

Source: Bordo, Humpage and Schwartz, Strained Relations. Calculations by the author.

Note: The table does not show US drawings from other countries.

The use of swap lines and the creation of the Gold Pool led to a warmer relationship between the Federal Reserve and the Bank. As described previously, in the 1950s the Bank had been reluctant to share information with the Fed when it came to foreign exchange intervention. The Bank also avoided using the services of the Federal Reserve for intervention in New York. The two institutions had always been in close contact, but the Bank had been unwilling to share too much information. This changed in the 1960s and even more so when the Bank started to use the swap network to support sterling. In 1964, the Bank had more telephone conversations with the Fed. It was more willing to share information on its intervention strategy and pass intervention orders to the Fed. The Fed even gained discretionary power on operations.⁹ The Bank would give the Federal Reserve a limit and leave operations to the discretion of the Fed operators. An extract of a telephone conversation memorandum best exemplifies this shift. Fousek, of the New York Fed, reported on a conversation with Preston of the Bank:

Preston called to say that he will be sending us an order to buy for their account up to £5 million at 2.7900. They hope they will have some beneficial effect from the announcement coming out of London (incomes policy). Should the spot rate move up, they would like us to use this opportunity and push up the 3 month forwards. For that reason, he is giving us a discretionary order for £10 million, value March 18, between 2.7714–25. If spot should start moving, depending on the situation, by moving up the forwards a little further, we might thereby get a beneficial effect for the spot rate. Should the spot really start moving, Blackler should be informed at home after 2:30 our time.¹⁰

In this telephone call the Bank delivered instructions for operations to be done overnight in New York. The Bank gave the Fed discretion. It gave permission for Blackler to be called at home if the situation became extremely worrying. In the 1950s the Federal Reserve could intervene only at the $2.78 lower band when things were going badly. In this quote, there is an order for $2.79 on the spot market. The Bank shared its strategy to try to benefit from good news and push the pound up. On the forward market the Federal Reserve was given discretion over £10 million. The goal was to improve the forward market with expected positive spill-over effects on the spot market. The spot market remained the Bank of England’s main focus. The total discretion of over £15 million is quite substantial, as the average daily intervention (including in New York and London) in 1964 was £6.4 million. The Federal Reserve had discretion to spend double the average daily intervention if it felt the situation required it.

This unique example is only anecdotal but is representative of changes in the daily telephone conversations between the two central banks. From 1962 to 1963 onwards cooperation strengthened, as reflected in many similar memoranda. The Federal Reserve was given both more discretion and more information about the intentions and activities of the Bank of England. Capie mentions that the 1964 Bank Rate rise was the first time that the Bank consulted the Fed in advance.¹¹

On a daily basis, agreements to swap currencies were made informally and quickly. For example, on 11 September 1964, Fousek reported that Blackler ‘informed [him] that he will be needing $15 or 20 million for Tuesday and will send us a cable on Monday. (This will be another swap drawing.)’¹² Fousek wrote in his note that he was ‘informed’ that the Bank needed swaps, giving the impression that such a request was never denied. Once the limit was agreed, the swap lines were an easy facility for the Bank of England to use. Amounts were agreed by telephone and confirmed via telegraph. It was easy for the Bank to obtain dollars within a day or sometimes less. The Federal Reserve noted in a 1964 memorandum that ‘undrawn amounts under swap arrangements with [the Bank of England, the German Federal Bank and the Netherlands Bank] may be considered available without prior consultation’.¹³ Swaps became one of the Bank’s preferred ways of accessing dollar credit.

Swaps were almost risk-free for the two parties. They included a collateral in currency backed by the other country. From the point of view of the Federal Reserve, the swaps were not only repayable in dollars and hence did not bear any currency risk. They were also backed by a collateral in pounds. If the Bank of England were to devalue while swap lines were still open, there was no risk to the Federal Reserve. The repayment would be in the original dollar amount. In the unlikely event of a British default, the Fed would still hold sterling as collateral. The low risk associated with swap contracts explains the low interest rates. Bordo et al. explain how interest payments worked: ‘The creditor central bank invested the foreign currency that it acquired from the debtor central bank for the term of the swap in a time deposit or in some other interest-earning asset. (The debtor would do likewise with any unused balances.)’¹⁴ In addition to bearing little interest, the unused balances could be reinvested, offsetting part of the interest payment.

Swaps between central banks were off-market accounting operations. They were a simple trade of IOUs. They had no impact on the foreign exchange market. They did not affect the money supply in either country opening a swap line or drawing on it (as long as the funds were not used). The Bank of England could then decide to use the money drawn from the swap line to intervene in the foreign exchange market. This would influence exchange rates and the money in circulation. Until the funds were used to intervene or pay a third party, swaps remained purely theoretical operations. They bore no consequences in the real world. This changed if the Bank decided to communicate the opening of a new swap line. This was the case in some rescue packages in the 1960s. The purpose of these rescue packages was to communicate to the market the willingness of the Bank of England to defend the pound. Otherwise, swaps were simply international reserves created ‘out of thin air’, as Coombs described it.¹⁵

Short-Term Swaps?

Swap contracts provided dollar liquidity to the Bank to defend the pound. They were designed as a short-term solution to temporary imbalances. Swap contracts were issued for three months but could be rolled over, as they often were. Table 8.2 highlights that, between 1964 and 1968, the Bank drew close to $1.5 billion on average every year, giving it a large dollar-denominated debt. This was not temporary, as the Bank constantly rolled over swap contracts. The press was aware of the short-term nature of the swap facilities. In 1963, the New York Federal Reserve reported comments in the British press: ‘[T]he agreement increases only short-term liquidity, and that it should be followed by greater Anglo-American cooperation in permanently increasing world liquidity.’¹⁶

Table 8.2. Annual summary of swap limits, drawings, repayments and outstanding debt

In million dollars	Swap line limit	Bank of England drawing	Bank of England Repayment	Outstanding debt to the Federal Reserve System
1962	50	0	0	0
1963	500	0	0	0
1964	750	1,370	1,170	200
1965	750	1,765	1,490	475
1966	1,350	625	750	350
1967	1,500	1,650	950	1,050
1968	2,000	2,045	1,945	1,150
1969	2,000	795	1,295	650
1970	2,000	400	1,050	0
1971	2,000	0	0	0
1972	2,000	0	0	0

Source: Bordo, Humpage and Schwartz, Strained Relations.

Table 8.2 demonstrates that the Bank of England had an outstanding swap position with the Federal Reserve in most years. From 1964 to 1969, the average outstanding was $646 million. During the same period, the Bank’s dollar reserves were $476 million on average and total reserves were $1,479 million on average. This meant that 43 per cent of the Bank’s reserves were short-term swap borrowings.¹⁷ Almost half of the reserves the Bank owned during that period were short-term US credit. Swaps thus came to be much more than a temporary liquidity facility. They were an inherent part of British reserves. The United Kingdom was treated favourably, as Table 8.1 highlights. It demonstrates the importance of sterling for the Federal Reserve.

The confusion on the maturity of swap credit lines was widespread. In 1966, even the UK government, in pretence or in fact, did not know that these instruments were temporary. In January 1966, James Callaghan, Chancellor of the Exchequer, met Coombs. According to a note, he ‘tended to assume that the Fed. swap could be rolled over beyond six months’.¹⁸ The Chancellor had only recently learnt that ‘the concept of the swap was that it should be for three months in the first instance with one extension of a further three months’.¹⁹ This highlights the privileged position of the United Kingdom when it came to swaps. The strategic importance of the pound meant that US policymakers, and especially Coombs, were extremely lenient when it came to dollar credit. The Chancellor’s surprise that these facilities were short-term reflects the extent to which these instruments gave the government breathing space. Efforts to improve the balance of payments and balance the budget could easily be deferred. There was an unlimited supply of cheap dollars. This meant that the United Kingdom had to make fewer sacrifices on fiscal and monetary policies than other countries. It was still able to maintain its Bretton Woods parity.

Swaps were more than a short-term liquidity solution for the United Kingdom. They were a feature of the Bretton Woods system. Did this enhance the stability of the international monetary system? Certainly, it helped avoid immediate crises. But did it improve the inherent stability of the Bretton Woods system? This is unclear. Contemporary observers noticed how it could increase the instability of the international monetary system. In a press review, the Federal Reserve noted how the UK Labour Party, before gaining power in 1964, was critical of swaps. The memorandum noted that the press was echoing ‘Labour’s reservation about the effectiveness of the [swap] agreement on the ground that it “placed us in a position in which two currencies, both liable to attack, were trying to support each other”’.²⁰ The dollar was inherently weak and prone to attack, as Triffin and others revealed. The dollar was then used to support sterling. In turn, a weak sterling could trigger attacks on the dollar. Swap networks certainly were a useful short-term fix, but they did little to increase the long-term stability of the international monetary system.

An Arsenal of Credit

Swaps with the Federal Reserve quickly became the Bank’s preferred credit instrument. Swaps were not communicated to the public, did not involve conditionality and could be agreed upon quickly and informally. The following paragraphs review the different sources of foreign exchange (mainly in dollars) available to the Bank of England. I present their advantages and drawbacks. Table 8.3 presents the sources of foreign currency at the disposal of the Bank of England. Privacy was a major concern for the Bank. It wanted to avoid communicating any losses or emergency loans unless they were significant enough to reassure the market.

Table 8.3. Schematic view of the Bank of England’s foreign currency credit instruments

	Own reserves	Federal Reserve swaps	BIS facilities or ‘Basel Arrangements’	IMF facilities (SBA, GAB and SDRs)	Private loans
Amount available (1960–71)	Up to reserve amount (average $1,773 million)	$50–$2,000 million	$200–$2,000 million	Up to $2,000 million	Up to third party’s willingness to lend
Conditionality	None	None	None	None in the early 1960s but progressive introduction of conditionality	None
Process	Internal	Telephone call and written confirmation	Request at Basel meeting or directly to members	Formal process	Private or through Eurodollar syndicate
Availability	Instantaneous	One day or less	Relatively quickly	Longer process	Relatively quickly
Public/private	Published every three months in the Quarterly Bulletin	Drawings completely private	Private or communicated when needed	Public or often disclosed/leaked to the public	Often leaked or public
Term	No term	Three months renewable	Short-term	Medium to long-term	Negotiable
Cost/interest	None or interest-bearing for US Treasury bills	Close to Treasury bills	Negotiated on ad hoc basis	1.5% per year on SDR	Market rates (usually higher than other forms of credit)

Credit sources in Table 8.3 are classified from left to right in order of increasing term. The short-term instruments are on the left, longer-term instruments on the right. The Bank had reserves that were immediately available. But they were both limited and published regularly (see Chapter 11). Publication meant that any change would be noted by the market. It had the potential to trigger a run on sterling. Using most of the Bank’s reserves, even to successfully defend sterling, was pointless as it would eventually bite back when reserves were published showing serious losses.

Swaps were favoured for their convenience, as we have just seen. They were one telephone call away and they offered total privacy. The Federal Reserve would communicate to the public only the limit of the swap line with the Bank of England. The amount drawn was secret. This meant that a swap could be raised without the market being informed.

The third dollar credit instrument was BIS agreements, otherwise known as Basel Arrangements. These were first used by Britain in March 1961 when sterling came under stress after a revaluation of the German mark and Dutch guilder of 5 per cent.²¹ It started as a short-term loan agreement. The loan was to be repaid by ‘the reflux of speculative funds or, if the reflux did not occur reasonably quickly, by recourse to the IMF’. These loan facilities were the last step before having to publicly apply for funds at the IMF. In addition to the Basel Arrangements, the United Kingdom also had access from 1965 to a Basel Group Arrangement. This special facility was intended to counter the effect of the conversion of sterling balances held overseas. The facility took the form of a swap agreement between the United Kingdom and the central banks of Austria, Belgium, Canada, Germany, Italy, Japan, the Netherlands, Sweden and Switzerland, as well as the BIS.

The IMF was the lender of last resort, in its most literal sense. The United Kingdom approached the institution only when strictly necessary, generally alongside other measures. The IMF offered various lending facilities that the Bank used: the Stand-By Arrangement (SBA) from 1952; the General Arrangement to Borrow (GAB) from 1961; and Special Drawing Rights (SDRs) from 1969. The SBA was available quite quickly and stood at $1 billion in 1964.²² The GAB offered a lending facility of $6 billion.²³ SDRs were created to expand international liquidity. They were one of the Bank’s least favourite credit tools. Because they involved a lengthy process, they could only be used as pre-emptive measures. The Bank had to request help from the IMF, which then needed to find a counter-party willing to provide dollars against SDRs. The counter-party was notified of the identity of the requestor. When used during a crisis, they would intensify the run against sterling by publicly admitting a position of weakness.

Private loans were at the disposal of the United Kingdom. They were not used frequently in the Bretton Woods period, but were used more frequently in the 1970s and 1980s. They were issued by banking syndicates on the Eurodollar market. The Eurodollar market is a dollar lending market located in London which emerged in the late 1960s.²⁴ In theory, these loans on the Eurodollar market had the advantage of being private and secret. However, in regard to the amounts borrowed, they were usually subscribed through a syndicate of private banks on the Eurodollar lending market. Schenk mentions the British government borrowing $2.5 billion from a syndicate including Chase Manhattan Bank in April 1974.²⁵ Schenk adds that public sector borrowers, including other countries, ‘raised $44.4 billion on the Eurobond market between 1963 and 1980’.²⁶

In 1964, Governor Cromer shared his intention to approach the private market for a loan with president of the New York Fed, Alfred Hayes:

Lord Cromer also mentioned the fact that Mr. John M. Meyer, Jr., Executive Vice President, Morgan Guaranty Trust Company, had just been visiting with him and proposing possible credit arrangement for the U.K. Lord Cromer had not of course informed Mr. Meyer of his earlier talks with Chase Manhattan. Lord Cromer and I agreed that while the credit idea had considerable merit, it would be well to keep it in abeyance at least for a little while longer.²⁷

It is noteworthy that the governor of the Bank of England openly discussed the matter with the Federal Reserve. The Fed at the time was another important creditor for the Bank.