4THIS TIME IS DIFFERENT?Crisis and Cooperation in the Twenty-First Century

Global Finance in the New Millennium

The crisis emerged in an economic and political context very distinct to the interwar and midcentury years. This was not a world of postwar reconstruction and rehabilitation, although the United States was engaged in its own wars in the Middle East, which shaped public spending, deficits, and debt. The means chosen to finance the US military buildup for the Iraq and Afghanistan wars fueled an economic boom and a housing bubble that eventually burst.⁵ But the aughts were not marred by reparations battles and efforts to rebuild the global economy after a global war.

The collapse of the Bretton Woods gold-dollar system in 1971 brought on the current fiat money system of mostly floating exchange rates. The US dollar remained the key global reserve currency. Access to dollar liquidity therefore remained a primary concern for the rest of the world. The system's functioning is shaped primarily by the United States and a handful of Western European economies. But the world had changed in ways that distinguish the contemporary era from the past. First, the arrival of a unified Europe with a common currency, the euro, and the European Central Bank. And second, the rise of new major players—China, India, Mexico, Brazil—has brought about a consequential shift in the global balance of power. Financial globalization and market integration had increased both interdependence and vulnerability of national economies to pressures overseas.

The global economic governance system, the rules that regulate the global economy, and the institutions that enforce them had evolved and expanded. Today, the task of large international financial institutions (IFIs) is to coordinate global action.⁶ Since the 1980s and 1990s, many central banks have gained de jure independence from governments to avoid short-term political interference in monetary policy decisions.

The central banking profession has also changed. John Singleton notes, “Central bankers in the early twentieth century knew little, if anything, about economics.” This could not be less true today.⁷ Central bankers today tend to be trained academic economists. How they engage with their counterparts, and how frequently they do so, has also evolved. They have moved away from the detailed and personal letter-writing of the past. However, technological advancements in travel and telecommunications mean that it no longer takes days or hours to communicate, but seconds. Central bankers are now in more frequent and instant contact than their predecessors.

Together, these changes make for a context very different to that of the gold standard or the Bretton Woods years, and were consequential for the outbreak, nature, and resolution of the GFC. Troubles in one part of the world, and especially in the United States, are even harder to contain to national borders today than in the past. International pressures called for an international response. Fortunately, the speed with which policymakers could communicate and the frequency with which they met facilitated rapid and coordinated responses from central banks as the crisis broke.

The Crisis Unfolds

Early in 2007, New Century Financial Corp., a company specialized in mortgage lending to cash-poor homebuyers in the United States, disclosed that many of its borrowers had started defaulting on payments. By March, its stock price had collapsed to $3.21 (from $15 earlier that year, which was already down by half its value). The company had declared it would stop making new loans, and it was now in need of emergency financing. In April, New Century Financial filed for bankruptcy. In the following months, large numbers of private-label mortgage-backed securities were downgraded to high risk and several subprime lending companies closed up shop. Dropping demand for housing fueled speculation of further house-price declines. Many subprime borrowers could not sell their homes to pay off their mortgages.

On August 9, 2007, a large French bank, BNP Paribas, suspended withdrawals from three investment funds that it managed. These funds were deeply invested in US securities markets, especially in mortgage-backed securities. In the next month, German banks faced collapse and were rescued. Bank failures in Europe peaked when Northern Rock PLC, a mortgage issuer in England, faced a cash crunch, as mortgage securities markets had become toxic. These bank failures were also associated with investments tied to the US subprime mortgage market.

As troubles emerged, central banks, governments, and monetary authorities were once again faced with a host of policy options to resolve immediate, local pressures and bank failures unilaterally, as well as through multilateral policy coordination.

The ECB was quick to intervene as a lender of last resort in the euro area. The Fed, eager to signal that it was on the same page as the ECB, released a statement that it was prepared to provide the necessary “liquidity to facilitate the orderly functioning of financial markets.”⁸ The central banks of Canada and Japan also intervened to manage liquidity pressures. The Fed cut interest rates just ten days after the BNP Paribas episode. On September 13, the Bank of England announced emergency support for Northern Rock, and the next day, depositors withdrew £1 billion, triggering the largest bank run in England in a century. Northern Rock was eventually nationalized, and the government took steps to guarantee depositors’ savings.

The problems faced by European banks soon returned to their place of origin. In Europe and the United States, large financial institutions such as UBS, Citigroup, and Merrill Lynch announced losses and exposure to bad debts, all associated with subprime mortgages. By the end of January 2008, a credit crunch had hit the world's richest countries and global stock markets faced their largest falls since September 11, 2001.

The crisis was punctuated with an onslaught of emergency financial packages and unilateral measures across national economies to rescue domestic banking sectors. These interventions came from central banks, with support from their governments.

Beginning in September 2007, the Fed and the ECB established a slew of bailout programs and began cutting interest rates dramatically. The Fed lowered the funds rate for its lending facilities by three-quarters of a percentage point to 3.5 percent, its largest rate cut in a quarter of a century. On March 7, 2008, it announced the release of $30 billion in a twenty-eight-day credit facility on March 10 and 24 through the Term Auction Facility (TAF), a new method to help sound banks acquire funds that they could lend on to consumers and businesses, which had been established in December 2007 and lasted through March 2010.⁹ This facility was established alongside the first round of central bank currency swaps with the ECB and the SNB.¹⁰ Until March 2008, the Fed tried to keep liquidity in financial markets, lending $200 billion to bail out bond dealers, who were stuck with toxic mortgage-backed securities and collateralized debt obligations.

The toppling of big banks culminated with Bear Stearns's collapse in the United States and was declared functionally bankrupt.¹¹ And although it was not regulated by the Fed and did not have access to the Fed's emergency assistance, its collapse would devastate financial markets. As a result, the Fed invoked its emergency authority under the Federal Reserve Act 13(3) to provide liquidity to “any individual, partnership, or corporation” in “unusual and exigent circumstances.”¹² It then further lowered interest rates to 2.75 percent, halving the Fed funds rate within a six-month period.

After what seemed like a lull in the crisis, and signs suggesting that the worst was over, troubles hit Fannie Mae and Freddie Mac, two federally backed mortgage companies, in mid-2008. Both institutions continued to suffer large losses and were placed into conservatorship amid the subprime mortgage crisis. Large British banks such as HSBC and HBOS revealed sinking profits, and house prices across the United Kingdom fell rapidly. The financial system that had been breaking down slowly now crumbled all at once. By the autumn of 2008, Lehman Brothers felt the pressures that had got to Bear Stearns some months before; on September 15, 2008, when the Fed had no legal way to hand over money to save Lehman, Lehman filed for bankruptcy.¹³ The same day, Merrill Lynch was purchased by Bank of America.

Amid the uncertainty, an excess of toxic assets and a systemic credit crunch, global dollar liquidity continued to collapse (see figure 4.1). Banks found it difficult to lend to one another. Worse, they could not lend to ordinary people.¹⁴ Banks’ assets were now considerably less valuable, if valuable at all, than when they had bought them. So, the Fed and US Treasury created a series of programs to strengthen domestic market stability and enhance market liquidity. That December, the Fed announced it was to lower its benchmark interest rate further to zero for the first time since the Great Depression. But, as central banks exhausted their monetary policy discretion, they could no longer lean solely on their go-to unilateral tools.

FIGURE 4.1. Global Liquidity Indicators (Quarterly), 2006–2014. Percentage year-on-year changes, in Euro, Yen, and US dollars. Source: BIS Statistics.

This was no ordinary credit crunch. The collapse of money and credit was systemic and had made its way abroad. Dollar funding strains became significant problems in Europe, the United Kingdom, and Japan. Norway had moved to protect its domestic banking system, while the highly leveraged Icelandic banking sector was inches from collapse. The problem was a lack of capital. They all needed help. For governments, arranging large packages was politically difficult to justify. And even had that not been the case, they would have still taken weeks to arrange. Across the world, national economies needed money to free up lending.

As the crisis worsened, Daniel Drezner argues that “despite initial shocks that were more severe than the 1929 financial crisis,” the global financial governance system of formal and informal rules and institutions “responded quickly and robustly.” In other words, the system worked.¹⁵ International organizations such as the International Monetary Fund (IMF), the Group of 20 (G20), and central banks were revitalized to respond in an “effective and nimble fashion.”¹⁶

Others argue that the role of this system is exaggerated—the IMF's resources fell short, and the G20 did little to bolster the Fund's capacity to meet the global demand for loans at the worst point of the crisis.¹⁷ Two of the most significant rescue measures emerged out of international cooperation between central banks. One is the first ever coordinated interest rate cut in October 2008. The Fed, the Bank of Canada, the ECB, the BoE, and the Sveriges Riksbank (in Sweden) all cut their primary lending rates by half a percentage point. The Swiss National Bank also cut rates. The Bank of Japan publicly endorsed the move. This coordinated rate cut gets the most attention as an act of unity among central bankers and made front-page news the day after it was announced. Despite this fanfare, the move had at best a modest effect on calming markets.

A more significant cooperative step was extension of liquidity swap lines by several major central bankers, led by the Fed. These steps were unprecedented in their amounts and their coordination.¹⁸ This program was neither an automatic nor an obvious policy response but rather symbolized the adage that situations of such gravity called for “outside the box” thinking.¹⁹

The largest swap network was established by the Fed, which extended bilateral currency swaps to a select fourteen central banks and is the focus of this chapter. The Fed essentially expanded the size of its balance sheet to deploy this program, and it was the institution's single largest crisis-fighting effort during the GFC.²⁰ The first lines were set up in December 2007 and continued through 2010, peaking with the largest swap amount of USD$170.93 billion to the ECB on October 2008. The Fed alone injected over half a trillion dollars of liquidity using these instruments. The Bank of Japan also extended dollar swaps to a few partner economies such as India or Indonesia, who experienced dollar shortages but did not receive a swap line from the Fed. South Korea received dollar swaps from both the Fed and the Bank of Japan. A euro network and a Swiss franc network also emerged. Of course, euro-denominated loans outside of the euro area were far fewer than those denominated in US dollars. Euro-denominated reserves made up about a fourth of world reserve holdings as opposed to two-thirds of world holdings being denominated in dollars.²¹ Poland, Hungary, and the ECB received SNB swaps.²² The People's Bank of China swap network emerged in 2009 and has since proliferated in its size and global reach. Using these swap lines, central banks injected trillions of dollars and euros into the global economy during the GFC between 2007 and 2010.

My study focuses on the swap network provided by the Fed, setting aside other central bank swap networks, such as those provided by the ECB or China. This is because, as was the case for the crises of the 1980s and 1990s, the kind of evidence necessary to illustrate and support my argument—interview accounts from recipients of ECB swaps, or real-time transcripts of the ECB's negotiations and decisions to extend these lines—as I provide in the Fed case, are unavailable.

At least seven central banks that requested a swap line from the Fed were denied one (see table 4.1). Even more, while these lines are largely viewed as unconditional, the terms of these arrangements vary between swap partners. A select few countries were given unconditional drawing authority, without further need for authorization from the Fed once a line was agreed to. For the four emerging market economies, limits were placed on the amounts and use of these instruments. Limits were presented as “additional safeguards,” and “lines could not be drawn on without further authorization, and individual drawings would be limited to $5 billion.”²³

These liquidity lifelines were arranged bilaterally between the swap-issuing central banks, here, with the Fed, and the recipient counterparty. Again, these lines emerged in an experimental and ad hoc manner and proved essential to bolstering the rescue measures undertaken by central banks and governments, providing critical measures to patch up the inadequate governance system. To many central bankers at the time, it was not initially clear that they would succeed.

By extending these liquidity lifelines, the Fed played the critical stabilizing role of ILLR. However, per Walter Bagehot's principle of “lending freely against good collateral at a penalty rate,” the Fed as an ILLR did not lend freely. Additionally, the collateral in a currency swap is the foreign currency being swapped, and the currency risk was assumed by the Fed's counterparty. But the risks of financial or currency collapse, and so the failure to honor the agreement, were systemic.²⁴ In other words, there was no good collateral. Now, central bankers’ discretion was paramount, especially in providing liquidity sans good collateral under conditions of uncertainty, and not risk, where the “probable distribution of outcomes itself is unknown.”²⁵ This is evident in the ambiguous, differentiated and arguably preferential application of the Fed's criteria in selecting swap recipients. Fed officials at the front line even recognized that the crisis was a period of radical uncertainty.

By mid-2007, before the worst of the crisis was upon them, FOMC officials cited concerns about unquantifiable uncertainty, not risk. In a June 2007 meeting, Jeffrey Lacker noted, “There is a vast range of uncertainty out there about which we can’t help markets and they can’t help us.”²⁶ In the September meeting that year, Donald Kohn, the Fed vice chair, found that the range of outcomes were “just too wide, and there's very little central tendency in it. So, I’d be very uncomfortable with a statement saying that I kind of thought the risks were balanced.”²⁷ By the end of the year, just as the Fed was about to extend its first swap lines, Kohn explicitly stated in a December FOMC conference call that financial institutions were protecting themselves “against a true Knightian uncertainty that they can’t price and don’t know how to protect themselves against.”²⁸ Fed officials found themselves in an environment of radical uncertainty as they entered the new year. Conventional metrics were no longer reliable, and policymakers had to rely on subjective reasoning as they worked to put out the fire.

Who had access to these swaps and who did not was certainly influenced in important ways by the structure and balance of economic power in the global financial system. Still, although these conventional accounts can explain our baseline expectations, the Fed made some puzzling choices as to who would or would not have access to these conditionality-free arrangements.

To explain the Fed's selection of swap recipients, the international political economy field turned to interstate and interest-group politics. Of course, interstate economic ties create an interdependence and thus credibility that actors will avoid defaulting on loans or cooperate to minimize mutual harm. And obviously, Fed officials had an incentive to assist foreign countries to protect the US banking sector and financial interests from overseas pressures; not doing so would imply that Fed leaders did not do their job.²⁹ Unsurprisingly, per our baseline expectations, a key predictor of receiving a Fed swap is the exposure of US banks to an overseas economy.³⁰

FIGURE 4.2. US Bank Exposure, 2007. This figure plots US bank exposure overseas for all countries who requested a swap, measured as the percentage of the consolidated claims of US banks in a foreign economy (in USD billions) as a proportion of total of the global consolidated claims of US banks as of December 2007. I exclude the UK and EU, which have the largest bank exposure, for space considerations.

Although US bank exposure and dollar shortages did motivate the Fed, a few cases stand out (see figure 4.2).³¹ Bank exposure in India was far greater than in several swap recipient states, while New Zealand, whose banking ties with the United States are among the lowest of all swaps requested, received a $25 billion swap. Among the fifteen economies facing the largest dollar shortages, such as Chile, Iceland, India, and Turkey, who requested swaps, were denied.³² Emergent market economies (EMEs) were undoubtedly disadvantaged in their international competitiveness and faced higher barriers to accessing this conditionality-free safety net.

We also know much less about how these lines came about and who was present at discussions and negotiations. As this book has shown so far, understanding the origins of such governance arrangements allows us to assess their robustness and their durability in the global financial safety net. In the rest of this chapter, I explain these outcomes discussed above and identify and illustrate the manner in which the Fed swap network emerged to highlight the interpersonal dynamics underlying this international cooperative response to the crisis.

The Inner Club

In 2007, a new cohort of central bank leaders found themselves at the front line—Bernanke, King, Trichet, and Shirakawa. In 2008, Newsweek ranked Bernanke, Trichet, and Shirakawa the fourth, fifth, and sixth most powerful people in the world who “could determine whether the global market avoids calamity.”³³ They entered the crisis with “different backgrounds, temperaments, and intellectual proclivities.” They did not agree on the causes of the crisis and held different views on how to resolve it.³⁴

Armed with the backing of their institutions, they took extraordinary, innovative, and improvised measures to tackle the crisis. They played crucial roles in the negotiations over international liquidity arrangements and managing financial instability at home and internationally. In the broader professional community of central bankers, a handful of central bank leaders were deeply enmeshed in the central banking fraternity; some were more deeply embedded in interpersonal ties than others. These relationships were integral to managing relations within central banks that today are far more decentralized than they had been in the past.

A central figure in this story is Bernanke, who himself is an expert on the Great Depression. His work with Mark Gertler explored the “expertise, information, and relationships” based on which banks made decisions to lend. Along with Simon Gilchrist, they later developed the concept of the “financial accelerator” that identified how economic shocks were transmitted through credit channels from the economy into the financial system in the 1930s.

In my interview with Bernanke, he noted, as he has also written elsewhere, that on assuming the Fed chairmanship, “an early priority was forging cordial working relationships with international policymakers.”³⁵ He already knew King when he took up the post. Bernanke and King both began their careers as academic economists. They first met when they shared an office suite at MIT in the 1980s and had a reunion lunch soon after Bernanke took office. While scholars would argue that this shared educational background lends itself easily to consensus and cooperation based on shared beliefs and preferences, that was not the case for Bernanke and King. Bernanke viewed the crisis as a “deeply intertwined set of risks to the banking system and the overall economy.”³⁶ King saw these problems and fault lines as less severe, viewing the crisis as a corrective for the long bout of banking excess.³⁷ Neither anticipated that they would be responsible for the world's most important currencies during the worst crisis since the Depression.

King also had a similarly old connection with Trichet, when they met at the other Cambridge—Cambridge University—decades before. Trichet was a lifelong bureaucrat who had studied Latin and the classics before beginning his civil service career. He “saw his profession as a central banker as being about something much bigger than economics.”³⁸ King and Trichet first met when Trichet was visiting Cambridge for his studies on the British tax system. Despite this old, academic connection, they too, like King and Bernanke, diverged on what caused the crisis and disagreed on the best approach to resolving it. Trichet initially viewed it as a one-off problem of banking panic and market uncertainty. And despite their divergent career and education backgrounds, Trichet and Bernanke saw it as essential to act quicky and assertively. King, on the other hand, saw regulating banks as messy and legalistic and found Trichet and Bernanke's initial response as an overreaction.³⁹

Certainly, international institutions, such as the Bank for International Settlements (BIS) or the IMF, provided central bankers a space in which they could meet and build their professional and personal relationships with their foreign counterparts. Bernanke first met Trichet in New York during an IMF meeting. This is also where he sought to forge relations with his counterpart in Mexico, Guillermo Ortiz Martínez, when he first assumed the chairmanship during his visit to New York for an IMF meeting.⁴⁰

Shirakawa, governor of the Bank of Japan, had been a part of the central banking world for decades and had spent time representing the Bank of Japan in New York earlier in his career, acting as general manager for the Americas in New York City. When he took up the governorship, he made the move from a faculty position in public policy at Kyoto University. Shirakawa is known for his unconventional and innovative ideas. But he also preached cautionary monetary policy and, throughout his term, resisted governmental pressure to take a more aggressive stance. He believed that a loose monetary stance would encourage “unchecked government spending and runaway inflation” and believed that it was the government's responsibility rather than the central banks’ to “encourage economic growth through structural reforms and other growth policies.”⁴¹

In interviews, several central bankers mentioned that they had developed close personal relationships of trust and goodwill with one another before the crisis through both their private and formal meetings and frequent correspondence and had developed long-standing personal familiarity with one another.⁴² Certainly, they did not share the intimate friendships of their interwar predecessors. But despite their different priorities and philosophies, they did share close relations of trust and goodwill, which helped them to converge on a plan to establish swap lines.⁴³ It was crucial that they could work together and have free and frank debates around their disagreements. It was also important that disagreement and criticism among central banks would only occur in private.⁴⁴ This was especially so for building trust that distinguished those central bankers who, like King or Bernanke, view themselves first and foremost as not politicians, from those with closer professional and personal ties to the political sphere.⁴⁵

Kohn noted that bankers today work within stricter institutional and legal constraints and do not share the deep and detailed letter-writing traditions of their predecessors. However, regular contact and correspondence allow them to build similar personal and social relationships with one another that give them added insight to their personalities.⁴⁶ Shirakawa also noted that technological advancement in communication has allowed them to maintain very private and regular contact both formally and informally.⁴⁷

Charlie Bean, then a deputy governor of the Bank of England, talked of the long-term personal relationships that some central bankers develop with one another over their careers.⁴⁸ But many central bankers noted that within the central banking club, there is an inner circle.⁴⁹ Others noted that institutions, especially the BIS, play a crucial part in fostering reciprocity and trust between central bankers.⁵⁰ The Global Economy Meeting and the BIS dinner allowed for a free and frank exchange of views over the goings-on in markets.⁵¹ The dinner is particularly important. Its closed, exclusive, and informal setting facilitates free-flowing conversation. The strongest interpersonal relations are cultivated in this confidential environment. Several central bankers noted that the exclusiveness of these dinners also generated strong personal ties between those who were regularly invited than those who often remained outsiders.⁵²

And although international financial institutions have routinized day-to-day cooperation and institutionalized intergovernmental credit facilities, once again, as in the past, the crisis decisively showed central bankers’ preferences for bilateral and ad hoc strategies to meet liquidity needs.⁵³ Institutions mattered in the GFC, but primarily in their role in providing opportunities for face-to-face interactions and communication among monetary authorities, such as at the BIS. But rather than turning to these long-standing institutional liquidity facilities, a handful of central banks sidestepped traditional avenues for intergovernmental lending for ad hoc, bilateral alternatives.

Unconditional Swaps

In a crisis, history counsels cooperation; its absence in the 1930s had disastrous consequences. During the GFC, advanced economies shed available strategies through the IMF or G7 accords to go their own way. Given that the crisis first erupted and spread across European and American banks, the ECB, the Bank of England, and the SNB were obvious recipients of Fed swaps. However, even among these more highly interconnected economies, the initial discussion and negotiations over swap lines were facilitated by favorable individuals and interpersonal ties that were conducive to early and rapid, ad hoc, and bilateral cooperation among these few banks.

In interviews, central bankers acknowledged and emphasized the importance of personal relationships and reciprocity among friends in facilitating such cooperation during the crisis, even though, as one interviewee said, “It's not the traditional point of emphasis for most researchers.”⁵⁴ Unlike epistemic communities, where policymakers need not meet formally or informally, in the context of cooperation during crises, it is important that individuals tasked with crisis management know one another personally.⁵⁵ It was also vital that they had opportunities to interact and discuss problems in private outside the public view. Of course, many of these individuals were also part of a broader and more diffuse community of economists, or central bankers, sometimes trained in similar traditions, and may have crossed paths in previous academic professions too. But these central bankers were not all career technocrats, nor did many of them share educational credentials. Instead, in the uncertainty of the crisis, relations of personal trust and reciprocity were essential in order to facilitate cooperation around new and experimental approaches to crisis management.⁵⁶

The origins of the Fed swap network were in a meaningful way grounded in relationships of trust and goodwill shared among a handful of leaders, which facilitated rapid bilateral cooperation early in the crisis. No doubt, most central bankers talked of the institutional and legal constraints on their discretion, the influence of domestic concerns, and the international economic and political climate on their foreign operations. But despite central banks’ large institutional apparatuses and operations within legal and institutional frameworks that outlive individuals, leaders’ influence increases in a crisis.⁵⁷

Swaps were typically arranged following a request from one central bank to another, usually the Fed. Agreements were reached over bilateral phone calls or emails, with personal appeals from governor to governor.⁵⁸ Initial conversations and informal agreements were made primarily at the leader level. Leaders often discussed these arrangements with their foreign counterparts prior to seeking their own banks’ support and approval.⁵⁹

Trichet referred to this cohort of central bankers as a “collective brain,” noting that the ties binding them together were not just professional but personal. He talked of how their easy and immediate personal relationships “of extraordinary confidence and intimate knowledge” were critical for generating collective solutions to the crisis, especially since they entered the crisis with differing views of both its causes and how best to resolve it. This is evident in that domestic crisis management policies varied across the world.⁶⁰ As central banks were exhausting their monetary policy discretion, one former central bank official suggested that in finding solutions to the crisis and making final decisions, leaders had the final word.⁶¹

Bean noted in an interview that while things do ultimately “get done,” under conditions of uncertainty, such facilities are difficult to rapidly get off the ground in the absence of personal relationships and trust.⁶² Indeed, as Jonathan Mercer notes, “If observers attribute cooperation to the environment rather than the person, then trust cannot—and need not—develop.”⁶³ Despite the ease and low cost of assembling swap lines, central bankers faced with the same crises have not been guaranteed the same privileged access to ad hoc liquidity assistance through swap lines. It soon became apparent that emerging economies were disadvantaged in their international competitiveness and faced higher barriers to accessing this conditionality-free safety net.⁶⁴

Generally, where central banks cooperated, it was not entirely surprising, but some cases can be identified where the necessary liquidity assistance did not occur. Requests from some central banks of systemically important or highly financialized economies, such as India and Iceland, were denied; others that did not unambiguously meet the Fed's stated criteria, such as Brazil or Mexico, were granted one.⁶⁵ Lending central banks suggested that liquidity shortages determined swap access. However, four of the fifteen economies facing dollar shortages—Chile, Iceland, India, and Turkey—that requested a Fed swap were denied.⁶⁶ Swap requests from the Dominican Republic, Indonesia, and Peru, which were not facing as severe dollar shortages, were also denied.

That the core, Western economies received the first unconditional swaps is not entirely surprising. These economies shared close banking and financial ties, and many were G7 members and fixtures at the BIS meetings and exclusive dinners. They interact regularly to incrementally build trust, goodwill, and familiarity over their careers.

Some non-US central bankers noted that Bernanke strongly favored showing a unified front among major central banks.⁶⁷ In previous studies, scholars have similarly emphasized Bernanke's pivotal role in pursuing the swap lines during the crisis in contrast to the nonaction or wrong action by central bankers in the 1930s.⁶⁸

How these swaps were arranged, however, is intriguing. Central bankers initially disagreed on how to manage the crisis, and such cooperation was not automatic. Before even getting to my questions, in an interview, the first thing that one former Fed official said was “You want to think about personal relationships. It is important that you can get your counterpart on the phone.”⁶⁹ They went on to say that, essentially, the earliest conversations and informal agreements were made privately at the leader level, often without knowledge of these discussions among their own central bank associates. Of course, they did indeed later consult with senior associates, who would be involved in executing these functions alongside partner banks.⁷⁰

But initial cooperation occurred outside traditional institutional avenues, through leaders’ interpersonal exchanges. When a solution was found, it was crucial that central bankers trusted one another to honor these agreements. The risks associated with the swap lines were mitigated by only extending them to trusted counterparties.⁷¹ When decisions had to be made quickly, institutional trust depended on interpersonal trust and reciprocity among leaders.⁷²

Some European bankers noted that Bernanke strongly favored showing a unified front among major central banks. Bernanke, an expert on the Great Depression, said to me that he had taken two technical lessons from that crisis. First, to not let the money supply collapse; and second, to ensure the availability of credit in the system. Second, he had also learned that in a powerful position, with powerful tools, a central banker ought to be “cautious, conservative and careful” in normal times. In contrast, in a crisis, it is often necessary to be bold and innovative, in consultation with colleagues and within political and legal constraints.⁷³ Caution and conservatism had prevented central bankers from managing the collapse of credit and the money supply a century ago. As discussed in the introduction of this book, Bernanke recognized the role of central bankers in the Depression and assured the audience of preventing a repeat.⁷⁴ Several of Bernanke's domestic and foreign colleagues who I interviewed suggested in interviews that this mantra undoubtedly informed his activist approach to the GFC.

Early on in the crisis, as discussed previously, central bankers were not unified in their agreement on how to manage it. So when a solution was found, as one central banker noted, it was crucial that central bankers knew and trusted one another to honor these agreements.⁷⁵ Much of the risk associated with these lines was mitigated by only extending them to trusted counterparties.⁷⁶ When decisions had to be made quickly, institutional trust depended on interpersonal trust and friendly relations among central bank leaders.⁷⁷ Early in the crisis, trust lay at the heart of the swap lines.

While some characterize the swap arrangements as unilateral defense mechanisms, motivated by the goal to protect US power and financial interests, they are bilateral arrangements.⁷⁸ Bernanke and his Fed associates depended as much on the trust, goodwill, and support of their counterparties as they did on the Fed. In many cases, not unlike Coombs's experience in the 1960s, Bernanke's counterparts initiated swap agreements even as pressures had eased.

In the summer of 2007, dollar funding pressures were growing in Europe; as David Wessel notes, European banks “couldn’t easily find [US dollars] in malfunctioning markets.” At the time, “key players at the Fed and at the ECB, for different reasons, were reluctant to make what seemed the obvious move,” which was extending swaps.⁷⁹

As pressures increased later in the year, it seems the tone was shifting. Interviews and FOMC transcripts show that the swap lines with central banks in Europe had been discussed and prearranged privately and informally between central bank leaders, prior to seeking their own banks’ approval.⁸⁰ In the FOMC meeting of September 18, 2007, on the question of dollar swaps and an auction facility, Bernanke shared that in “some conversations [he] had, in particular with President Trichet of the ECB, [they] came up with the possibility of combining these two things, essentially having auctions simultaneously in the United States and Europe, and then using the swap markets to provide the dollars to the extent that the ECB would like to have them.” They had arrived at a similar plan with the Swiss National Bank.⁸¹

Across the Atlantic, European central bankers, initially resistant to the swap proposal, later expressed, as Adam Tooze writes, that they “didn’t expect to have any difficulty getting hold of dollars” from the Fed.⁸² ECB officials noted that Trichet and Bernanke had discussed these arrangements privately. Both leaders then had to gather the support of their banks, which they would generate in internal meetings to get consensus support, so as to not go back on their word. The internal deliberations over predetermined settlements were therefore essentially pro forma, as central bank leaders had already informally agreed to these arrangements, grounded in interpersonal trust and reciprocity.⁸³ By making these agreements informally, leaders had great discretion over their banks’ decisions.

Unlike Bernanke and Trichet, King viewed the crisis as a corrective for banking excess and was eager to avoid a moral hazard problem from bailing out banks. Earlier in 2007, he strongly opposed any intervention for liquidity support as he believed it “encourages excessive risk-taking and sows the seeds of future financial crises,” which explains why the Bank of England did not join its European counterparts in proposing swap lines with the Fed.⁸⁴

But when talking about the lead-up to and decisions made to arrange a swap, King emphasized people and personal relationships between central bankers mattered. He was close with Kohn's predecessor, Roger Ferguson Jr., with whom he had arranged a dollar swap with the Fed on September 12 and 13, 2001, immediately following the attacks on the World Trade Center in New York City. King suggested that the 2007–2008 swaps emerged in a similar manner. In an interview, when I asked King how these lines first came about, he said, “If you trust your counterpart, for temporary and emergency purposes, you can give the Fed a call.”⁸⁵

Paul Tucker, who assumed the deputy governorship at the Bank of England in 2009, was close with Kohn at the Fed. He suggested that friendly and informal relations with Kohn, developed largely in Basel, helped facilitate discussions on how to collectively address liquidity pressures. Kohn similarly said that he and Tucker “had developed a close working relationship that was very useful when [their] economic and financial systems were under stress.”⁸⁶

The possibility of coordinated action was broached in broad terms between Tucker and Kohn in South Africa when walking to a formal dinner. Tucker sensed an appetite in the Fed to take joint action. They had a very long, free, and frank conversation (“there's something about a good walk and talk”), where they could discuss the need for coordination or cooperation in ways that would demonstrate “joinedupness” to the world while each central bank tailored its actions to its own particular circumstances. This personal, informal discussion between Tucker and Kohn helped lay the groundwork for official follow-up discussions between the Fed, Bank of England, and others. When, a few weeks later, a handful of central banks announced new measures, the Financial Times's front-page headline focused on the coordination.⁸⁷

The first lines were extended in December 2007. By September 2008, the Fed's swap network encompassed the ECB, the Bank of England, the Swiss National Bank, and the Scandinavian central banks.⁸⁸ And in September 2008, when Japan sought out a swap, following an extraordinary meeting in the middle of the night, Shirakawa notes that “the arrangement was up and running in a matter of a few days, without any information leaks before its launch. It reflected a strong mutual trust among central banks.”⁸⁹ When recounting this cooperative effort in an interview, Trichet claimed that the collective brain was “equipped with the appropriate synapses,” opening the possibility for an early, quick, and cooperative response to the crisis.⁹⁰

The crisis generated unusual circumstances for central bankers. With conference calls occurring at unusual hours, as Shirakawa writes, “to negotiate and hammer out practical solutions with overseas counterparts, sometimes without sleep,” knowing people well, and knowing how to talk to, negotiate, and reason with them was essential.⁹¹ King talked of his extensive interaction with Kohn, with whom a US-UK swap had already been discussed informally. Interestingly, these plans for collective efforts between the United States and United Kingdom ran counter to King's initial refusal to lend to the system.⁹² But central banks could not rely on governments for large amounts of liquidity, nor could they arrange swaps publicly. Confidentiality and trust were crucial, and reciprocity and personal relationships mattered fundamentally for cooperation and as traits that are a mark of central bankers who work closely together.⁹³

Some central bankers note that in the earliest days, these huge agreements that central bankers entered into, often overnight, were arranged solely on the word of central banks’ leaders.⁹⁴ Given the uncertainty around the success of crisis management efforts, the earliest and largest swaps between major central banks were grounded in their governors’ personal trust. Almost nothing about their outcomes was known ahead of time. In his memoir about the crisis and its aftermath, Bernanke recounts his worry that early swaps may not have sufficed in Europe: “This may not work. I don’t want to oversell it,” [Bernanke] told the FOMC. “If we do it, we are just going to have to give it a try and see what happens.”⁹⁵ This worry and uncertainty around the success of these policies was echoed by several central bankers.⁹⁶ Through these swap lines, the Fed's lending to central banks reached almost USD$600 billion between 2007 and 2010, peaking at $170.93 billion in an overnight swap to the ECB on October 15, 2008.⁹⁷ When later asked by Alan Grayson, a Florida Democrat, during a congressional hearing, which foreign banks were lent the money by the Fed, Bernanke answered, “I don’t know.”⁹⁸

Since the announcement of these lines, the use of these tools has been seen as extremely problematic in the public and political spheres. Gerald O’Driscoll, formally of the Dallas Fed, called out the New York Fed for undertaking this effort in a manner that is “troublesome in a democracy” and for using its authority “to bail out European banks.”⁹⁹ Others have criticized the Fed for extending swap lines during the crisis that were issued without either approval or oversight from Congress and from the White House.¹⁰⁰ Even those who acknowledge the Fed's need to play the role of an international lender of last resort are concerned with the potential risks and public costs associated with these arrangements in their current form.¹⁰¹ And, by 2010, the FOMC itself came to recognize the political risks associated with these measures, especially, as Michelle Smith, the Fed's communications director warned, anything that they were “keeping secret” that would fuel speculation in Congress.¹⁰² But in its earliest days, keeping these efforts behind closed doors, between just a few, apolitical central bankers, was essential.

Neil Irwin writes that the Fed's extensive but “hidden liquidity support measures” were “a closely guarded secret even by the standards of the always secretive Fed.” During the panic, “information about the Fed's lending was so closely held—and had it been known publicly, so potentially explosive—that only two people at each of the dozen reserve banks were allowed to access it.”¹⁰³ The FOMC meetings in which swap lines and other programs were finalized were followed by “closed meetings” with the Board, which published no transcript, only “very summary” minutes released two weeks after the meetings. The information provided was “very general—something like there was a discussion of means to address money market issues.”¹⁰⁴

Throughout the crisis, central bankers had taken to innovation and improvisation, somewhat departing from convention of caution in favor of what Bernanke and others refer to as “blue sky thinking.”¹⁰⁵ Central bank staff had been working on a wide range of proposals that were circulated to Bernanke, Kohn, Geithner, and Kevin Warsh. Deliberations over these “pre-decisions” took place in closed meetings. Fed Board meetings are subject to “sunshine laws”: formally, a meeting is made public and informative, which Bernanke observed in his book, was “not a great venue for blue-sky thinking and strategizing.” And some interviewees noted that the Fed's legal team could make exceptions as to what constituted a meeting.¹⁰⁶

Smaller meetings of fewer than four Board members did not trigger this open-meeting law, and because Geithner was not a Board member at the time, these meetings were exempt from the sunshine requirement. Because such a setting was more suitable for blue-sky thinking, these group meetings were often kept exclusive to these four individuals, while Bernanke tried to keep other Board members not in attendance at these informal meetings “apprised of developments through one-on-one lunches and frequent email exchanges.”¹⁰⁷

The Bank of England also chose not to announce major policy decisions and actions, not only to the public, but to politicians as well. Arranging swaps and the coordinated rate cut would have been “impossible to do publicly” given their immediate impact on markets and the widespread effects of interest rate changes.¹⁰⁸ Other central bankers also alluded to the ambiguity over what counted as a meeting in their banks—whether three people in an elevator or conference calls with foreign counterparts was a meeting was debatable.¹⁰⁹ But it was these private, closed door environments that created the space for sensitive decision-making, especially where expectations could impinge on the markets.

The first swap agreements expired in February 2010 just as liquidity pressures slowly began to ease. When pressures mounted in Greece early that year, the Fed did not have a swap line in place with the ECB despite the possibility of contagion via European banks.¹¹⁰ But the Fed did not rush to reactivate or enhance them. The Eurozone crisis had not quite begun; problems of drying up liquidity in the United Kingdom and Japan eased. Still, Trichet, King, Shirakawa, and Bernanke sought to reopen their swap lines that had just expired. These steps could not have been taken in the absence of interpersonal connections between Bernanke and the rest.

In an impromptu conference call that interrupted FOMC officials’ Mother's Day celebrations on May 9, 2010, Bernanke broached the idea of reopening these swaps with the ECB, the Bank of England, and the Bank of Japan: “Yesterday Jean-Claude Trichet called me and made what I would characterize as a personal appeal to reopen the swaps that we had before. This morning I have gotten, again, personal calls from Mervyn King, of the Bank of England, and Masaaki Shirakawa, of the Bank of Japan, also asking us to reopen the swaps.”¹¹¹

Bernanke went on to say that this exchange with Trichet followed an ECB Executive Board meeting where they came to “very significant decisions” that Trichet shared with Bernanke on “a highly confidential basis.”¹¹² In acknowledging the extraordinariness of these steps, Bernanke reemphasized that Trichet's call was a “personal appeal” and that Trichet “feels it is very important for us to support him, and he understands [the Fed's] concerns.”

King and Shirakawa's appeals carried a similar personal tone.¹¹³ And in interviews, where I discussed this moment in the crisis with King and Shirakawa, they both echoed Trichet and Bernanke's sentiment. Shirakawa noted that at the time, the sentiment expressed by the Fed was that there was no need for a swap; on the ground, Japan was not affected by the emerging European crisis. To request this new swap from the Fed, it was helpful that Shirakawa had a close personal relationship with Bernanke, which made broaching the topic and agreeing to reopen the lines much simpler.¹¹⁴ King's account of this decision was similar, adding that these central bank leaders’ reciprocal and cooperative manner of working was facilitated by goodwill and interpersonal trust between himself and Bernanke.¹¹⁵

In sum, interpersonal ties of trust and reciprocity played a crucial role in facilitating bilateral and ad hoc cooperation, even among the core Western economies during the crisis. These relationships were especially important in renegotiating these lines when there was no longer a need for them. Moreover, the swap network emerged through private, informal, and personal discussions during the crisis, rather than formal, institutional channels.

Differentiated Ties

In 2008, four EMEs also benefited from US liquidity assistance: Mexico, Brazil, Korea, and Singapore were granted a Fed swap in 2008. With the exception of Mexico, this was the first time that emerging markets were included among the Fed's swap recipients. Fed officials reasoned that these are four of the largest and systemically important emerging economies. Each of these central banks was offered up to $30 billion for three months.¹²⁰ Studies show that the Fed's criteria in selecting swap recipients were not applied unambiguously to the emerging markets that requested a swap to access this conditionality-free option.¹²¹ Instead, geopolitics and diplomatic ties facilitated this cooperation.

Although some suggest that these instruments are used to exert leverage against risky partners, I argue that these geopolitical motivations are not motivated by leverage but by favoritism: by assisting those with whom Fed leaders shared close interpersonal relations of trust and goodwill.¹²² Fed leadership mitigated substantial risks incurred in these agreements by excluding those with whom they did not share strong personal ties of trust and goodwill and assisting only those central bank leaders they did trust, to show a united front.

By focusing on the case of Mexico, I show that arranging a swap agreement with EME central banks relied on central bankers’ relationships of interpersonal trust, goodwill, and reciprocity. But where relational ties differed, so did access to dollar liquidity. Conversely, the absence of such personal affinities between bank leaders hindered India's chances at receiving a Fed swap. India matched Mexico and Brazil on the Fed's criteria for a swap, especially regarding the Fed's key justification for their swaps of having a GDP of over $1 trillion. Yet it was unsuccessful in its bid to acquire a Fed swap during the crisis.

A Standing Bazooka

In 2008, Bernanke was keen to get temporary authorization to use swap lines during the crisis, “so I know I’ve got my own bazooka here.” And the crisis demonstrated that this bazooka had proved vital to the global crisis-fighting efforts. And so, in a 2009 FOMC meeting, Sheets presented a memo proposing the establishment of standing swap lines with the Bank of England, ECB, Bank of Japan, and the Swiss National Bank. These standing lines essentially now provided these partner central banks with a more permanent backstop to dollar markets abroad. Technically, the Fed could still prevent the use of these lines, but now, these large financial centers would be less reliant on ad hoc arrangements than they had been in 2007 and 2008, without having to go through the difficulty of renegotiating these lines in the event of a crisis.

Here, I provide new details on how the Fed deliberated and came to establish five standing swap lines with the United Kingdom, Canada, Japan, Switzerland, and the ECB. The focus here is less on the interpersonal aspects of this decision-making, as the FOMC transcripts suggest that the push for these permanent lines largely came from the Fed itself and not from a partner's request. However, deliberations in the Fed's initial conversations in 2009, and the eventual authorization of these permanent lines in 2013, highlight two important political- and individual-level dynamics: First, the Fed recognized that where it drew its line was a sensitive, political question, which had implications for its relations with partner banks. Second, Sheets was no longer at the Fed in 2013 when the FOMC established these standing lines, and the considerations of who ought to be included in this group changed, with the notable absence of Mexico.

The “standing” swap lines proposed by Sheets in 2009 would be limited in their amounts and in partner banks’ abilities to draw on them. These lines would effectively be reauthorized annually when the FOMC is reconstituted each year.¹⁶⁸ Sheets proposed including Canada and Mexico's lines through NAFA in this group, as “not enhancing the lines might strain [the Fed's] relationships with them.”¹⁶⁹ In response, James Bullard, then president of the St. Louis Fed, asked about the “difficulty of having to, as you say, renegotiate and set up the details, as you did this time,” for discontinued lines, and the implicit likelihood that these partners may not get a swap in the future.

Sheets's response to Bullard shows that central bankers thought about the individuals involved in negotiating these lines making decisions, and consequently about how personnel changes could affect crisis management in the next crisis. Sheets was less worried about scaling up a program than of the difficulty of going “from zero to four” lines, and not from four to fourteen “if the world blows up … eight or ten years out.” Why? Because “ten years from now, the folks in New York who did all the work may be onto other things or in other positions, and just as was the case in this last episode, we had to run around and pretty much figure these things out from scratch.”

Sheets also questioned the assumption that countries that had received a swap in the GFC ought to get a temporary swap again, saying, “Now, the flip side of this is that we gave them a swap line once, and if things get bad, we’re going to give them a swap line again. I think that is a very reasonable presumption if a crisis erupts in a year or two years. But if the next crisis is eight or ten years out, there is a lot of water under the bridge. The world looks a lot different, and it's not clear to me that there is the same presumption.” It may end up that the FOMC, in the event of this later crisis, may choose to give the same partners another swap line, but he did “not think there would be as strong a presumption as if they actually had a swap line with us.”

Sheets wanted to arrange standing lines only with the United Kingdom, Europe, Switzerland, Japan, and potentially Canada and Mexico. To him, there was a qualitative difference among these six parties, in terms of their economic size, and in the case of Mexico and Canada, their proximity to the United States. On that point, Bernanke joked, “The North American location is hard to change.” Janet Yellen, then president of the San Francisco Fed, and later Fed chair, and Eric Rosengren, then president of the Boston Fed, pushed for a larger set of standing swaps.

The FOMC debate illustrates the committee's concerns about managing their relations with partner banks, stigmatizing those that were not granted standing swaps, and geopolitics. Rosengren worried about the unintended consequences and the risk that the Fed's reasons for limiting these standing lines might be misinterpreted. A swap line with Mexico would imply that the Fed was “comfortable having them as a counterparty.” By not expanding the network, Rosengren was concerned “about stigmatizing the countries like Mexico and Korea and other countries that want potentially to be a counterparty in part for the signalling benefit.” Yellen agreed and was concerned about the economic consequence of excluded parties’ potential reactions if the Fed then told, say, “Korea … well, we are not putting you in the same category as before.”

FOMC members found it was necessary to delineate those in and outside this standing network by some clear metric. But what that metric would be was a sensitive question. “Size” would pose problems for justifying these limits to exclude Asia, as Bullard observed: “When I say Swiss, ‘It's a small country, come on. This is an old club that you guys have been fostering for years.’ And ‘You just don’t like us because we’re in Asia.’ I can imagine that that is sort of the attitude.”

No decisions were made in 2009. But the FOMC recognized the utility of having standing lines, that “coordinating policy decisions with foreign central banks has the potential to be complicated and time consuming, so it's preferable to have as many as possible of these decisions negotiated in advance.” They also noted that “the swap lines have been a powerful symbol of central bank cooperation.” This proposal was discussed occasionally in the years that followed and five standing lines were established in 2013, with the Bank of Canada, the Bank of England, the Bank of Japan, the European Central Bank, and the Swiss National Bank.

There is little evidence of how the Fed's negotiations with its counterparts went in 2013, but the FOMC discussion to create these permanent lines that year was quick and notably different to the previous one. Specifically, Sheets, who had been a key architect of the GFC swap lines, especially in emerging market lines, had left his position at the Fed. Relatedly, the current discussion was whether to convert temporary lines with the Bank of Canada, the Bank of England, the Bank of Japan, the European Central Bank, and the Swiss National Bank, into permanent ones. Mexico, whose 2008 swap had expired, was no longer included in conversations alongside Canada. In 2013, FOMC members were more willing to differentiate between Mexico and Canada without including the NAFA lines that both countries had.

Some FOMC members wanted to at least inform their Mexican counterparts ahead of time about the Fed's decision to establish its standing lines. Fisher expressed concern that Mexico was not included and not given advanced notice of this decision. Some other FOMC officials, however, noted that it was not typical to provide counterparty banks any advance notice about these decisions, in the October 2013 FOMC meeting. Fisher noted, “They’re highly sensitive, and, again, it is part of NAFTA, and with any differentiation between Canada and Mexico, I would suggest politely that we let them know at the earliest opportunity, and before others, if possible.”¹⁷⁰ But barring that minor debate, the discussion around this standing facility was short. Five permanent lines with the Fed were unanimously approved in October 2013, without including Mexico, as Sheets had initially proposed in 2009. These lines proved vital in 2020 with the outbreak of the COVID-19 pandemic, as I will discuss in the conclusion.

What's Past Is Prologue

Even though central bankers today do not operate with the same free rein of their 1920s counterparts, when the worst crises emerge, the discretion and decision-making power of individual leaders and high-level personnel is substantially heightened to facilitate quick action to arrest emerging troubles. Bernanke, King, Shirakawa, Trichet, and their international and domestic associates walked in the shoes of Strong, Norman, and their international colleagues by responding to the GFC with extensive intervention to provide liquidity in the drying up global economy. In periods of crisis and uncertainty, especially when time is short, and the trajectory of current problems is unknown, central bank leaders lean on a range of heuristics and personal judgments in their decision-making. These assessments are shaped not only by institutional relations and economic signals but also by individuals’ personal relations of trust, goodwill, and familiarity with their counterparts.

With each passing decade, central banks, governments, and other policymakers have invested endless resources and countless hours in creating, building, and strengthening the rules and institutions that govern the global economy with the goal of establishing a robust and reliable global financial safety net. And still, when the global economy faced an imminent crisis in 2007, monetary authorities and central banks sidestepped this system to take the ad hoc route.

Insofar as monetary authorities avoided a second Great Depression in 2008, yes, the system worked. But as the crises emerged and escalated from 2007, just like their predecessors experienced in the 1920s and the 1960s, central bankers once again took to ad hockery and governance innovation to stem liquidity pressures around the globe. While these tools had been used before, it was still unclear whether they would work at all this time or how far they would go to mitigate the crisis.

Once again, the crisis was resolved by a handful of central bankers who decidedly sidestepped this system of preexisting and institutional forms of intergovernmental lending and opted for a more ad hoc approach to crisis management. In the absence of this trust, the personal appeals that central bankers made to their friends in the United States may not have had the same sway. Many of their initial discussions around arranging swaps, and agreements over these decisions that were made informally, over dinners or on long walks, may not have been possible.