However, when events are not anticipated, fear in the money markets can quickly emerge. And these fears are invariably reflected in LIBOR. For instance, on 29 November 1999, the one-month US dollar LIBOR jumped 86 basis points because of fears surrounding the so-called ‘millennium bug’. Considering that the Federal Reserve often tended to cut or raise interest rates in 25 basis points at a time, and rarely changed them more than a few times per year, 86 basis points represented a remarkably large move. Banks had spent billions on system upgrades and contingency plans relating to the Y2K software problem. What would happen if we arrived in the office after the New Year’s celebrations to find that the date was 1 January 1900, rather than 1 January 2000? A collapse of the banking system and a return to the Stone Age, it was thought. Therefore, banks borrowed money as a precaution and LIBOR shot up.
Soon after I had returned from Tokyo to London and joined Crédit Agricole Indosuez, two planes flew into the World Trade Center in New York. As George W. Bush reportedly headed to Camp David, banks hoarded cash. However, as the Federal Reserve cut rates as a response, the three-month US dollar LIBOR fell 75 basis points within ten days. The effect of 9/11 lasted for months and years. The fear of sudden events that could cause a meltdown of the financial system became incorporated into trading. But the trading did not stop. Rather, traders grew accustomed to how to react to crisis situations, and elaborate trading strategies, often involving LIBOR derivatives, were crafted to take advantage of them. Hurricane Katrina, the London 7/7 bombings and many more became market-moving ‘events’, as they created uncertainty, panic and ultimately mispricing of LIBOR-indexed derivatives.
Importantly, both market and liquidity risk can change without having any impact on the perceived credit risk of the banks, the second component of the money market risk premium. However, liquidity issues can also be closely related to credit issues. If the perceived credit risk of a bank is high, the bank should find it more difficult to borrow money. Therefore, it would have to pay a higher rate to compensate for this and LIBOR should go up. The bank might then find itself actively seeking to raise cash, and at the same time trying to reduce lending, as a precautionary measure. However, traders not active in the Japanese yen market (or having no memory of the banking crisis almost a decade earlier) became used to very small deviations of LIBOR from what could be regarded as the risk-free interest rate. Banks did not go bust, and they were not expected to do so any time soon. Access to liquidity was easy, and central banks became increasingly transparent and predictable in what they were doing. Overall, LIBOR (and its equivalent benchmarks elsewhere) seemed to be working as intended. Although the market had become more sophisticated, many traders and banks believed that the new normal would last forever. A situation in which banks would have to pay a considerable premium above the current and expected future central bank rates to access liquidity was seen as unthinkable. A situation in which European banks would have to pay a premium to access US dollars was seen as impossible. As long as the banks’ risk management systems showed that there was no risk, there was an illusion that no risk could exist.
***
When I received a phone call from the Merrill Lynch office on 9 August 2007, I was in the middle of chopping wood in the Swedish countryside. Although I trusted my assistant who covered for me whenever I took a few days off, I always had my mobile with me just in case.
It can sometimes be difficult to connect to a trader when you are ‘off’. The buzz in the dealing room is contagious, and traders often become hyper without any particular reason. Sometimes, when you are not there, you almost have to pump yourself up to get a certain level of heartbeat in order to have a meaningful conversation. But this time it was different. I repeatedly had to ask him to calm down as he kept on stuttering about FX swaps, dollars, prices people had dealt at – without making any sense at all. He kept repeating that things were ‘crazy’ and ‘completely mad’ out there. There was not much I could do about it from a woodshed more than 1,000 miles away, so I asked him instead whether our trading positions were OK.
‘It’s difficult to say,’ he said, ‘but I think we’re OK.’
During our phone conversation, it became clear to me that something quite extraordinary was happening in the financial markets. Something I had never seen or heard of before. At the time, there was nothing in what he said that made me worried about my position, let alone the global financial system. However, it prompted me to buy an earlier Ryanair flight back to London. I had to see it with my own eyes. I needed to get back into the game.
As a trader, you regularly have to change your opinion. The market changes every second and new information constantly feeds your brain. Often, you have to accept that you were wrong, take the loss, forgive yourself and move on. However, some trading ideas might be more long term. They can become fundamental to how you see the market, and such convictions can stay with you not for days or weeks but for years. They become part of your ‘trading style’ and shape the person you are as a trader. I had several trading styles, certainly, but one of them was that I normally thought that short-term interest rates would go up rather than down. Having been trained in the aftermath of the Scandinavian banking crisis, and then having experienced the Japanese banking crisis, I always had an underlying fear that another banking crisis loomed around the corner. Consequently, I tended to ensure that my long-term trading position took that into account. This view often deviated from the market consensus and was sometimes both irrational and foolish.
I did not, of course, predict the global financial crisis. I did, however, ensure that my trading book would generate money should the global financial system tip over into disaster. I also happened to be working for a bank that had a reputation for taking risks, and for encouraging traders to do so too. I generally took a lot of risk, and risk taking as a banking activity had increased exponentially since I had become a trader. The net outcome was that my trading positions were enormous. Overall, I had put bets on LIBORs, on the barometers of fear. Bets that a vicious storm would come in.
The world did not end on 9 August 2007. But my trading positions were definitively more than just ‘OK’. They were magnificent.
***
A little over a year later – on Monday 15 September 2008, to be precise – I went to work as normal. Having exited Farringdon tube station, I walked through Smithfield market towards King Edward Street. It was about 6.30 a.m., and trading in the biggest meat market in the country was about to finish for the day. Trucks had already been loaded with fresh meat and the cleaners had begun to spray the surrounding area, as they always did at this time of day. The blood had to be cleaned up before the commuters arrived.
When entering the Merrill Lynch Financial Centre, I took the escalator, ordered a triple espresso from the in-house Starbucks and went into the dealing room. This was not going to be a normal day. People often talk about the financial markets being ‘24/7’ or ‘around the clock’, but this is not entirely true. Traders rarely go into the dealing room at the weekend. When the lights go off in San Francisco on Friday, the market does not properly open until Wellington wakes up on Monday morning. The day before, on Sunday 14 September, I had been called into the office. The reason was Lehman Brothers. People no longer feared that the US investment bank might go under. We now knew they were going under.
When I got into the dealing room, it was as if someone had died. But this was not a sudden or unexpected death; rather, it was a slow and painful process that had reached an inevitable outcome. No matter how fierce the competition between Merrill Lynch, Lehman Brothers, Morgan Stanley, Goldman Sachs and Bear Stearns had been, there remained a sense of respect and togetherness among the US investment banks. Everything would change from that day on. With Lehman Brothers declared bankrupt, Merrill Lynch would not have survived another day had Bank of America not stepped in and rescued us at the very last minute. Surprisingly, perhaps, the traders in the dealing room did not seem thankful.
Even
today, central bankers, regulators and other policy makers talk about measures that ought to be taken to avoid another ‘Lehman moment’. What they refer to is a situation in which the entire global financial system is on the brink of total collapse – when you stare into the abyss and realise that the world is about the re-enter the Stone Age. The trigger, but hardly the cause, was the demise of the US sub-prime mortgage market during the first half of 2007, which had a snowball effect. At first, some asset-backed securities markets that hardly anyone had heard of ran into trouble. The virus then spread to markets that had traditionally been seen as relatively safe. The crisis kept on coming closer to traders such as myself, and soon substantial losses were reported by names we were familiar with: the UBS hedge fund Dillon Read, two hedge funds run by Bear Stearns, and the US home loan lender Countrywide.4 The money market began to dry up and did so quickly. Then, when the German bank IKB reported that they had rollover problems, we knew that the crisis had spread from the US across the Atlantic. More and more hedge funds reported losses, forcing them to sell assets to raise cash and to post more collateral to their brokers. Everyone, it seemed, needed to borrow money. Nobody, however, dared lend any. On 9 August 2007, the French bank BNP Paribas announced that they were to freeze redemptions for three investment funds, citing its inability to value complex structured products. Although an analyst had claimed to Bloomberg that it wasn’t ‘too significant’ (considering how large the bank was) and that it was ‘more of an image problem’, I was not so optimistic. BNP was a ‘proper’ bank. IKB was not a player, and many had never heard of Countrywide. Everyone, however, had heard of BNP. They had always been part of the trading community and had made markets in everything: FX, bonds and derivatives. They were active in major currencies and in emerging markets. They had branches everywhere and were also French. The French banks were famous for hiring the best programmers and mathematicians. If they did not know how to value their books, who did?
The central banks reacted fairly quickly by pumping money into the banking systems. The European Central Bank (ECB) injected €95 billion and the Federal Reserve $24 billion. A week later, the Federal Reserve went a bit further by broadening the type of collateral they accepted for lending to the banks. They also lowered the ‘discount window’ (the level at which banks were allowed to borrow from them) by 50 basis points and increased the lending horizon to 30 days. However, the measure was not deemed a success. The 7,000 or so banks that were based in the US and could borrow at the discount window from the Federal Reserve were reluctant to do so because of the stigma associated with it. Using the discount window would signal desperation and hence a lack of creditworthiness in the eyes of the rest of the market. Nobody wanted to look as if they were in trouble. From that standpoint it was indeed an ‘image problem’.
The problems refused to go away. October and November 2007 saw a series of write-downs of sub-prime and other mortgage-related assets. The total losses kept being revised upwards. When the Federal Reserve realised that the interest rate cuts announced during the autumn were not filtering through to the money markets, it introduced the Term Auction Facility (TAF), an arrangement whereby US-based banks could borrow from the Federal Reserve without using the discount window.
Similar market movements were observed in other currencies, with central banks across the developed world resorting to comparable measures. Central banks found themselves in a difficult position as the symmetry of the monetary transmission mechanism had broken down. Price stability through inflation targeting had gradually become more important than financial stability as a central bank goal. However, the former goal no longer applied. Having become more transparent themselves, central banks now had to rely on information and signals provided by the banks and the markets. The key indicator, the ‘LIBOR–OIS spread’, provided evidence of severe stress in a range of currencies and markets. The overnight index swap (OIS) was a derivative instrument that was indexed to the ‘risk-free’ central bank interest rate. The idea was to reduce the LIBOR–OIS spreads towards levels seen before the crisis broke out. From August 2007, the LIBOR–OIS spread became the focal point for everything we did as traders. This is also why, in 2009, the former Chairman of the Federal Reserve, Alan Greenspan, famously described the LIBOR–OIS spread as ‘the barometer of fears of bank insolvency’.5 It was precisely that: a kind of fear index relating to banks. This fear had soared since 9 August 2007.
From a personal perspective, the global financial crisis also acted as a trigger point in revealing the wider implications of LIBOR.
The STIRT desks at the banks had not been particularly glamorous before 2007. We traded a range of instruments that were important, but not interesting and complex enough to represent the forefront of financial innovation. The turnover in OISs, FRAs, IRSs (interest rate swaps), CRSs (cross-currency basis swaps) and FX swaps, among others, was enormous. However, options, long-end interest rate swaps and structured products enjoyed considerably more prestige. The crisis turned everything upside down. Suddenly, the spotlight fell on us. Options traders needed to know the direction of LIBOR in order to price customer deals and value their books correctly. So did the bond and long-end swap traders. FX traders needed to know where they could fund the currencies they had sold, and the same applied to the cross-currency basis swap traders. As everything we did had a direct link to the interbank money market and LIBOR, the STIRT desks evolved from being merely profit centres to a place people turned to in order to get some ‘market colour’.
Which bank was rumoured to be in the worst shape?
Which bank seemed most desperate to borrow?
Which bank had been ordered to throw in the towel and temporarily suspend their market-making activity?
Which bank had refused to deal with which other bank for fear of insolvency?
All of these aspects, coupled with continuous reliable and less reliable information flows in the dealing room and in the corridors about how our bank was coping, forced senior management to take notice. As interest in LIBOR and the markets closely connected to it grew, it did not take long before central bankers joined in on the conversations. I had spoken to numerous central bankers before, probably representing a dozen countries. The discussion topics had generally related to the technicalities of new derivative instruments, overall market liquidity, and the arrival (or departure) of new banks and large clients in the market. Naturally, they were interested in whether the market in their currency was functioning well. As a trader, it was like being called into the principal’s office to provide an update of what was happening in the school playground. A phone call from a central bank gave me an immense sense of pride, a feeling that what I did for a living was important. Now, central banks seemed interested in only one thing: LIBOR.
In a way, this sudden change was not surprising; until August 2007, LIBOR had more or less appeared to work as intended, namely as a largely harmonious outcome of the central bank on the one hand, and various market participants on the other. Seen from this perspective, our job had been to implement their policy: to ensure that the first stage of the monetary transmission mechanism worked as intended. They obviously wanted to get this mechanism working again.
However, it was the phrasing of the questions I got from the central banks that surprised me. One senior central banker was under the illusion that LIBOR had to be correct because how could a price on a Reuters screen be wrong? A central banker representing another country believed that their LIBOR was traded on the stock exchange. A third, and very senior, central banker seemed to know roughly how the process worked, but privately said that the soaring LIBOR–OIS spreads were a good thing as they acted to dampen inflationary pressures in their economy.
I was shocked by how little they seemed to know about the benchmark rate they now wanted to get under control. But in truth, I was not wholly surprised. For years I had thought that central banks were much less powerful than the public was led to believe. ‘Central banks are put at a constant dis
advantage versus the market when it comes to implementing monetary policy,’ I scribbled down on a piece of paper in early 2009. ‘The LIBOR problem has implications as it delays information to policy makers who are supposed to steer LIBOR. This probably led to a very long delay in the rate-setting process after the credit crunch started in 2007,’ I then went on to write as I tried to formulate a research question for my PhD application.
There was nothing wrong with the way central bankers saw the market, in theory. LIBOR should reflect the rate at which banks lent to each other, so the idea that the LIBOR–OIS spread was a kind of barometer of fear of bank insolvency was logical. The standard technique they then seemed to be using was to measure this fear in detail by quantifying each of the components that made up the LIBOR–OIS spread. By assuming that LIBOR was a perfect reflection of the money market, and taking the OIS market prices as representing the risk-free interest rate for a given maturity (i.e. the final payment date of the financial contract), it simply became a task of allocating the difference between the two variables into the appropriate credit and/or liquidity components making up the spread. In fact, if a measure for credit risk could be agreed upon, the remaining component could be regarded as liquidity risk. This was the approach taken by the Bank of England6 in an indicative decomposition of LIBOR. Following this approach, liquidity risk (effectively the fear of not being able to get hold of cash), rather than credit risk (the fear of default), was shown to have been the main driver behind the widening LIBOR–OIS spreads since the beginning of the global financial crisis. Central bank action, aimed at reducing this spread through extensive liquidity injections, confirmed this thought process. Similar studies were done by other central banks, showing the same results and all pointing to the same conclusions.
Barometer of Fear Page 5