The Foreign Exchange Matrix. Barbara Rockefeller

Harding, ‘Application of Random Matrix Theory to Economics, Finance and Political Science’, Department of Economics, MIT Institute for Quantitative Social Science, Harvard University, SEA’06 MIT: 12 July 2006 (web.mit.edu/sea06/agenda/talks/Harding.pdf). [return to text]

      2 Robert Slater, Soros, The Unauthorized Biography, The Life, Times and Trading Secrets of the World’s Greatest Investor (McGraw-Hill, 1995). [return to text]

      3 James Burke, The Pinball Effect (Little Brown, 1996). [return to text]

      Chapter 2 – Review of Risks

      “If this were a logical world, men would ride sidesaddle.”

      Ron Frost, Agricultural Markets Manager, Chicago Mercantile Exchange

      Markets are increasingly global in scope, so that a Mumbai investor can trade a US equity issue or the euro/dollar exchange rate with equal ease of execution (although we may question whether he should). A symptom of this globalisation is that global financial markets move together in a way that they did not before; events in one market have repercussions for others.

      There are many risk indicators that players watch to try to help them understand the risk that movements in one market will affect another they are interested in, but we do not yet have a single risk index to explain market behaviour.

      In this chapter, after discussing the globalisation of FX markets and contagion, we move on to the tools the market uses to gauge risk. We discuss VIX, the highly-watched fear-factor gauge, risk reversals, stress indexes, and credit default swaps.

      Risk and contagion

      Contagion arises from the Asian Crisis

      Before the Asian Crisis of 1997-98, a big risk event was fairly well contained to the country in question; now a Brazilian limit on foreign investor inflows has a spillover effect not only on the Mexican peso, but also the Korean won. The Asian Crisis was the first time we saw the attitude toward risk shifting from a single country basis to a global basis. What began as a run on the Thai baht in May 1997 (leading to devaluation in July 1997) spread to investor selling of the Malaysian ringgit, Philippine peso and Indonesian rupiah, among others. In the autumn of 1997 and early 1998, Asian currency jitters sent shock waves around the world, with other emerging market currencies feeling the heat as investors exited existing emerging market long positions and sought safe-haven currencies.

      By May 1998, the Russian stock and bond market had collapsed, along with the Russian ruble. Russian officials were forced to triple interest rates to 150% to prevent the ruble from falling further. In June of that year, the Japanese yen succumbed to downward pressure, with the Bank of Japan and the US Federal Reserve intervening to prevent further yen weakness. The dollar-yen topped out at ¥147.65 before ending the year at ¥113.45.

      By August 1998, the Dow Jones Industrial Average had fallen over 500 points and, by September, Federal Reserve Chairman Greenspan promised to lower US interest rates as the Nikkei 225 hit a 14-year low and other global stock indices also fell dramatically. In addition, troubled US hedge fund Long-Term Capital Management (LTCM) received a $3.5 billion bailout. LTCM had been counting on the convergence spread in fixed income to make tiny arbitrage gains magnified by high leverage, a strategy designed by not one, but two, Nobel Prize winners. Like others, they had never heard of contagion on this scale.

      In the autumn of 2008, the Fed initiated a series of interest rate cuts to stabilise US financial markets. The International Monetary Fund (IMF) announced bailout packages for Russia and Brazil. The market began to stabilise in the first quarter of 1999 and the Dow Jones Industrial Average climbed back to over 10,000. Investors breathed the first sigh of relief in nearly two years. While traumatic, the events of that period provided investors with a valuable education. They learned the hard way that, as New York Times columnist Thomas Friedman would later write, “The World is Flat.”

      Globalisation of risk

      While we were not looking, economic globalisation had occurred far faster than anyone was accounting for. No longer can risk be confined to the properties of a specific security and its environmental conditions, but an investor has to look at risk on a worldwide basis. In the first decade of the 2000s, you couldn’t open a newspaper without seeing reference to the butterfly effect, whereby a butterfly flapping its wings in China results in a tornado in Kansas. Note that the formal name for the butterfly effect is sensitive dependence on initial conditions, giving rise to just about any observation about an economy becoming expandable into grand deductions about markets near and far.

      Since the Asian Crisis, when a big risk event occurs, market players judge whether the risks will be contained to one country or region or are global in nature, and react accordingly. Investors’ collective attitude towards risk becomes a factor in its own right and influences how markets react to shocks. As a report of the Bank for International Settlements (BIS) put it:

      “Bad news in a market situation where investor risk appetite is already low is likely to result in a much greater repricing of risky assets than in periods where it is high. The dynamic stance of the risk appetite of market participants as a sentiment could thus serve as an important contributing factor in the transmission of shocks through the financial system. Furthermore, as it might itself be influenced by the situation in financial markets, it could work as a multiplier. Accordingly, taking into account the risk appetite/risk aversion of investors and its evolution has become an important element of assessing the condition and stability of financial markets.” ^[4]

      In recent years, larger risk aversion on a global scale has been triggered most often by events taking place in the United States. The market panicked after the NASDAQ Composite crash in 2000, in the wake of the 9/11 attack of the World Trade Center, in the lead-up to the Iraq War in 2003, following Hurricanes Katrina and Rita in 2005, and during the subprime mortgage crisis that began to unfold in late 2008.

      In contrast, the euro zone, which came officially into existence in January 1999, was the root cause of fewer worldwide risk events until recently. Small effects in the euro/dollar exchange rate were occasioned by the French and Dutch No vote on the referendum in favour of the European Union’s proposed constitution in 2005 and Ireland’s rejection of the Lisbon Treaty in 2008 (later accepted in the 2009 vote), but these events were contained to Europe. Having previously avoided being the source of a shock catalyst, Europe has more than made up for it with the peripheral country debt crisis that began in the fall of 2009, with Greece admitting it had cooked the budget books on the Olympics. This set off a chain of developments that has ended up endangering the structure and composition of the euro zone itself.

      Asia had a chance to offer a shock again, in the form of the Shanghai Surprise from October 2007 to October 2008. The SSE Shanghai Composite Index was already falling from a peak in October 2007 when (on 28 February 2008) it fell 9% in a single day. The S&P fell the next day by the most since 9/11/01. European and Japanese markets fell. Certain other emerging market stock markets, like Brazil, Russia and Turkey, also suffered big declines – but not all emerging markets.

      The Shanghai event was curious for many reasons, not least that at the time the exchange had limited international participation and the drop was triggered by reports of impending new restriction on transactions, including a proposed tax. By what logic does this isolated market – and why do other regional markets – have such an outsized effect, such as is illustrated in Figure 2.1, globally?

      Figure 2.1 – the Shanghai Surprise (SSE Shanghai Composite Index (Dark)) vs. S&P 500, Nikkei 225 and FTSE 100

      Explaining large-scale contagion

      Analysts don’t have a good explanation for large-scale contagion. To call it a Pavlovian or herd-instinct response is both insulting to the investor class and insufficient. Why should one