Do you remember February 5, the day in which volatility spiked and the now-well-known XIV ETN collapsed after losing 94 per cent of its value? This also caused the US stock market to suffer a 10 per cent slump and a global lost in the equity market of $4.2tn, more than the losses suffered by the Nasdaq index after the dotcom bubble burst. The one of the last February was a great event, a very rare one; someone might call it aBlack Swan, a metaphor invented by Nassim Nicholas Taleb to describe an event regarded as almost impossible.
Indeed, this was a great event because it did remind us that in finance, “almost impossible” things actually do happen and that there is no such a thing as free-lunch.
A free-lunch is, in broad sense, a strategy in which an investor is able to earn a profit without taking any risk. This is what people that invested in the XIV thought they were doing: the idea that, if “nothing bad occurred for a long period, then things cannot go much bad” had shaped the mind of such investors. But the recent happenings proved them wrong.
In this article, I would like to exploit the recently high market volatility to try to better explain the concepts of risk and its related measures and to highlight how the market participants tried to address the market risk issue.
Over the past six decades, volatility has come to dominate risk-management models across the finance industry while, at same time, investment bankers have tried to create securities allowing investor to buy and sell volatility, just like any asset. Such new securities made the market more “complete” but at same time created a potentially dangerous feedback loop thus making booms and busts even bigger in magnitude.
The academic history of volatility starts in the 1950s with Harry Markowitz. At the time, people already had the intuitive idea that riskier investments should generate higher returns to compensate the danger of losing money, but there was no academic approach to address that. Then a 25-year-old Markowitz came and wrote its masterpiece: “Portfolio Selection”. Published in the Journal of Finance in 1952, it argued that returns should be judged against, and optimised for, the amount of risk taken. In order to identify in a more rigorous manner the broad concept of risk, Markowitz decided to use the “variance”, or volatility, as a handy proxy.
While Markowitz was not the first to use volatility as a proxy for risk, he was the first to put it in a rigorous framework. What he actually did was to introduce the idea of “optimising for risk”, meaning that for a given return on a given investment, an investor should minimize the volatility of such returns in order to maximize his/her utility: this was the first step into the so-called “modern portfolio theory”. Today this seems pretty obvious if not trivial, but such idea won Markowitz a Nobel prize in economics in 1990. Right after Markowitz studies, there has been another economist (and a Markowitz’s student) who continued to pursue the idea to use volatility as a proxy for risk and to convince the market that that was the right choice: William Sharpe.
Sharpe has been a brilliant professor at the University of Washington. There, in 1966, he published a paper named “Mutual Fund Performance” in which he introduced the now well-known “Sharpe ratio”, a simple ratio providing a proxy for the reward to variability. In other words, directly comparing the returns of a fund manager to the volatility of his performance, and subtracting the returns of a risk-free asset such as cash. Such measure is so simple to understand and to implement that almost every fund manager includes the Sharpe ratio in its investor prospectus in order to provide to the investors a clear measure of the riskiness of the product they hold. Despite such academic background came to life in 1950, it took time for the market to get used to volatility-as-risk. Back in the 1960s and 1970s, accurate financial data were not easily available and the computers used on Wall Street were unable to properly calculate the volatility of various markets or stocks.
Therefore, it was not so immediate for market participants to implement such straightforward idea.
Things started to change in the light of the 1987 Black Monday crash, when JPMorgan’s then chairman Sir Dennis Weatherstone ordered staff to start a daily report which briefly revealed how the bank exposure on its trading positions on any given day. The “value-at-risk” (or VaR) was born. Such report was handled on Sir Dennis’s desk at 4.15pm every day and exploited historical volatility of markets to calculate the maximum loss the bank could suffer with a high degree of certainty. The Risk metrics, as the JP Morgan model is known, had a few competitors, but it was the only one to be widely implemented in the industry. The dominance of VaR and its dependence on volatility as a proxy for risk became crucial during the financial crisis, when it became clear that models were not able to properly describe empirical distribution of losses. This is because VAR model assumes that market returns (hence losses) are normally distributed which in many cases proved to be a good approximation. However, during specialevents, returns distribute according to a PDF with “fatter tails” than the standard normal PDF. The inability of the VAR to model Black Swan events, as we called it at the beginning of this article, represents the main flaw of the parametric approach as it yields lower estimates of losses. Also, the use of the volatility in VAR can boost “pro-cyclical” behaviour: during low volatility markets, investors should buy more assets, thus lowering volatility. On the other hand, when volatility start rising, risk models suggest to underweight assets, thus triggering sales on the market place and boosting volatility even more.
But this is just a part of the story. Even as volatility became the dominant way to measure and manage risk, a series of academics and investment bankers started laying the groundwork for trading volatility itself. This turned volatility from being just an observable phenomenon into something that investors could themselves influence by trading it.
So do investor trade volatility? Let’s start from the basics. Volatility is a key factor in explaining derivatives pricing. All such products involve an implicit bet on volatility: buying a put option, i.e. the right to sell the underlying at a fixed price at a fixed point in time, which is like buying insurance against price declines. On the other hand, buying a call optionis a way to bet that a stock will rise. A combination of such securities allow traders to pursue different strategies: by buying a call and a put at the same time, for instance, they can form a “straddle”, which in practical terms means betting on market turbulence. A bet on market tranquillity is achieved by selling such position. Also, is important to remember that options owe their popularity to Fischer Black, Robert C. Merton and Myron Scholes, three academics who, in 1973, published a revolutionary model providing a closed-formula to compute options price. The iconic Black-Scholes model won Merton and Scholes the Nobel prize for economics in 1997 (Black had passed away a few years before). This can be regarded as the first big step towards making volatility a tradable asset. Then, in the 1980s Menachem Brenner and Dan Galai published a series of papers and created Sigma, an index of stock-market volatility based on options. They took the idea to various exchanges but at the time no one was really interested into a live volatility benchmark. The idea was then recollected in 1992 by Robert Whaley, an employee of the Chicago Board Options Exchange (Cboe). By 1993, the Cboe Volatility Index was born under the name of Vix.
So what’s Vix? It is an index aiming at measuring the expected volatility of the US stock market over the next 30 days, as implied by option prices which theoretically means measuring the level of investor anxiety. Due to the way it is calculated, which involves thousands of underlying derivatives, investors couldn’t trade the Vix itself. At least not yet. The first “pure” volatility derivative is thought to be a deal structured by a UBS banker called Michael Weber in 1993. Weber, created the first “variance swap”, a contract currently traded on a daily basis on the markets. It was based on the UK stock market’s volatility and its purpose was to protect the Swiss bank’s trading book from losses. Variance swaps quickly started to gain ground on Wall Street, especially in the late 1990s, when the Asian financial crisis hit and the collapse of hedge fund LTCM heavily shook the markets. This rose interest in the idea of trading volatility itself, but variance swaps remained too sophisticated for wide usage, at least until the beginning of the new century.
Despite the strong interest, the Vix has become a traded index only more recently: it summer 2002. A few years earlier, Mark Cuban had sold Broadcast.com, an internet radio company, to Yahoo for a cool $5.7bn, leaving him with $1.4bn worth of the portal’s stock. Then the dotcom bubble burst and Cuban wanted to buy some protection. He asked to a Goldman Sachs banker who tried to sell Cuban a variance swap, but the Texas-based billionaire wanted to get exposure to the “fear index” he had heard so much about. Soon afterwards he mentioned the businessman’s desire to punt on the Vix to a colleague, Sandy Rattray, which was stroke by the idea and immediately contacted the CBOE to make a proposal.Cboe hired the investment bank to develop Vix’s methodology and to make tradable futures contracts linked to it available on the market. In 2004, the first Vix futures was there. At the very beginning, interest was restricted to a few group of investors. But when the financial crisis erupted, the large demand for insurance on the market made the Vix index vary popular in the finance industry. People rushed to buy Vix futures to protect themselves against the heavily bearish market, and by 2008 more than 4,300 contracts were traded on a daily basis.
However, this is not all the story. Financial engineers soon opened the Vix market also to retail investors by constructing exchange-traded products based on the index. In 2009, Barclays built the first volatility-linked ETP using Vix futures, and by 2017 there were more than 40 Vix-linked ETPs available to trade. Among them, some are “long volatility” and some are “short” with an average daily trading volumes around 294,000 contracts last year. But, as February’s turmoil made clear, it very risky to have a relevant input into risk-management models that is also a popular trading tool.
Then, we get to the last February.
Launched in 2010, the Global Volatility Summit, an annual meeting of hedge funds dedicated to trading turmoil, has never been as largely and intensively attended as it was this year. The money manager pointed out the worrying implications of just $3bn of Vix-linked ETNs causing such market turmoil, highlighting how volatility had transformed from a proxy of risk into an input for risk, comparing it to the tail-eating uroboros snake of Ancient Greece mythology. During such event, Christopher Cole, founder of Artemis Capital Partners, warned that February’s events turbulence were just an appetiser for a “volatility revolution” that “will not be televised”. Cole alleged that there is now a “dangerous feedback loop linking volatility, low interest rates and financial engineering, and estimates that there is more than $2tn in strategies that both exert influence over, and are influenced by, stock market volatility”, a journalist at FT reported. “This has made volatility the only asset class that really matters”. Although XIV was of small matter, compared to the whole market it ended up being the spring that triggered a panic attack. Indeed, banks and investors hit by its collapse, strove to hedge their positions, in turn triggering a vast amount of automated selling by the volatility-targeting funds. Luckily, the pretty favourable economic environment smoothed the impact of such chaos, encouraging many investors to take advantage of the sell-off and gain from price bouncing. Cole added that the economy can’t be so buoyant every time such a shock happens and that the “XIV-triggered vol-mageddon was therefore just an amuse-bouche for what the future might hold”. “I don’t know if I’m crazy, or if the rest of the world is crazy,” he added. “But this is the kind of thing our children might look back and say, how did we not see this coming?” None has the right answer to such complicated questions. Nonetheless, people who have played a part in the history of volatility defend their role in its evolution while expressing concerns at the next stage. Also, volatility is now embedded in risk-management models. Trading strategies have been built on volatility targets. Making volatility a tradable asset has been, on the one hand, a valuable development, but, on the other hand, instruments such as the Vix ETNs proved to be very harmful. In a few years, our children might look back and say, how did we not see this coming?
Christopher Cole Physicists have long noted that observing some phenomena actually changes their nature. In finance a similar rule applies. As more people weave volatility into their models, they change its nature in subtle but important ways. But the point is: can or even should anything be done about it? Also, is it really so that the volatility-trading industry is big enough to shake the $80tn global stock market? Despite all the analysis and questions about what should or should not be done, Markowitz, the founder of volatility as proxy of risk, is sceptical that anything will ever change. “It has organised so much of our world that there’s simply no way of backing away from it,” he alleged.
Indeed, this was a great event because it did remind us that in finance, “almost impossible” things actually do happen and that there is no such a thing as free-lunch.
A free-lunch is, in broad sense, a strategy in which an investor is able to earn a profit without taking any risk. This is what people that invested in the XIV thought they were doing: the idea that, if “nothing bad occurred for a long period, then things cannot go much bad” had shaped the mind of such investors. But the recent happenings proved them wrong.
In this article, I would like to exploit the recently high market volatility to try to better explain the concepts of risk and its related measures and to highlight how the market participants tried to address the market risk issue.
Over the past six decades, volatility has come to dominate risk-management models across the finance industry while, at same time, investment bankers have tried to create securities allowing investor to buy and sell volatility, just like any asset. Such new securities made the market more “complete” but at same time created a potentially dangerous feedback loop thus making booms and busts even bigger in magnitude.
The academic history of volatility starts in the 1950s with Harry Markowitz. At the time, people already had the intuitive idea that riskier investments should generate higher returns to compensate the danger of losing money, but there was no academic approach to address that. Then a 25-year-old Markowitz came and wrote its masterpiece: “Portfolio Selection”. Published in the Journal of Finance in 1952, it argued that returns should be judged against, and optimised for, the amount of risk taken. In order to identify in a more rigorous manner the broad concept of risk, Markowitz decided to use the “variance”, or volatility, as a handy proxy.
While Markowitz was not the first to use volatility as a proxy for risk, he was the first to put it in a rigorous framework. What he actually did was to introduce the idea of “optimising for risk”, meaning that for a given return on a given investment, an investor should minimize the volatility of such returns in order to maximize his/her utility: this was the first step into the so-called “modern portfolio theory”. Today this seems pretty obvious if not trivial, but such idea won Markowitz a Nobel prize in economics in 1990. Right after Markowitz studies, there has been another economist (and a Markowitz’s student) who continued to pursue the idea to use volatility as a proxy for risk and to convince the market that that was the right choice: William Sharpe.
Sharpe has been a brilliant professor at the University of Washington. There, in 1966, he published a paper named “Mutual Fund Performance” in which he introduced the now well-known “Sharpe ratio”, a simple ratio providing a proxy for the reward to variability. In other words, directly comparing the returns of a fund manager to the volatility of his performance, and subtracting the returns of a risk-free asset such as cash. Such measure is so simple to understand and to implement that almost every fund manager includes the Sharpe ratio in its investor prospectus in order to provide to the investors a clear measure of the riskiness of the product they hold. Despite such academic background came to life in 1950, it took time for the market to get used to volatility-as-risk. Back in the 1960s and 1970s, accurate financial data were not easily available and the computers used on Wall Street were unable to properly calculate the volatility of various markets or stocks.
Therefore, it was not so immediate for market participants to implement such straightforward idea.
Things started to change in the light of the 1987 Black Monday crash, when JPMorgan’s then chairman Sir Dennis Weatherstone ordered staff to start a daily report which briefly revealed how the bank exposure on its trading positions on any given day. The “value-at-risk” (or VaR) was born. Such report was handled on Sir Dennis’s desk at 4.15pm every day and exploited historical volatility of markets to calculate the maximum loss the bank could suffer with a high degree of certainty. The Risk metrics, as the JP Morgan model is known, had a few competitors, but it was the only one to be widely implemented in the industry. The dominance of VaR and its dependence on volatility as a proxy for risk became crucial during the financial crisis, when it became clear that models were not able to properly describe empirical distribution of losses. This is because VAR model assumes that market returns (hence losses) are normally distributed which in many cases proved to be a good approximation. However, during specialevents, returns distribute according to a PDF with “fatter tails” than the standard normal PDF. The inability of the VAR to model Black Swan events, as we called it at the beginning of this article, represents the main flaw of the parametric approach as it yields lower estimates of losses. Also, the use of the volatility in VAR can boost “pro-cyclical” behaviour: during low volatility markets, investors should buy more assets, thus lowering volatility. On the other hand, when volatility start rising, risk models suggest to underweight assets, thus triggering sales on the market place and boosting volatility even more.
But this is just a part of the story. Even as volatility became the dominant way to measure and manage risk, a series of academics and investment bankers started laying the groundwork for trading volatility itself. This turned volatility from being just an observable phenomenon into something that investors could themselves influence by trading it.
So do investor trade volatility? Let’s start from the basics. Volatility is a key factor in explaining derivatives pricing. All such products involve an implicit bet on volatility: buying a put option, i.e. the right to sell the underlying at a fixed price at a fixed point in time, which is like buying insurance against price declines. On the other hand, buying a call optionis a way to bet that a stock will rise. A combination of such securities allow traders to pursue different strategies: by buying a call and a put at the same time, for instance, they can form a “straddle”, which in practical terms means betting on market turbulence. A bet on market tranquillity is achieved by selling such position. Also, is important to remember that options owe their popularity to Fischer Black, Robert C. Merton and Myron Scholes, three academics who, in 1973, published a revolutionary model providing a closed-formula to compute options price. The iconic Black-Scholes model won Merton and Scholes the Nobel prize for economics in 1997 (Black had passed away a few years before). This can be regarded as the first big step towards making volatility a tradable asset. Then, in the 1980s Menachem Brenner and Dan Galai published a series of papers and created Sigma, an index of stock-market volatility based on options. They took the idea to various exchanges but at the time no one was really interested into a live volatility benchmark. The idea was then recollected in 1992 by Robert Whaley, an employee of the Chicago Board Options Exchange (Cboe). By 1993, the Cboe Volatility Index was born under the name of Vix.
So what’s Vix? It is an index aiming at measuring the expected volatility of the US stock market over the next 30 days, as implied by option prices which theoretically means measuring the level of investor anxiety. Due to the way it is calculated, which involves thousands of underlying derivatives, investors couldn’t trade the Vix itself. At least not yet. The first “pure” volatility derivative is thought to be a deal structured by a UBS banker called Michael Weber in 1993. Weber, created the first “variance swap”, a contract currently traded on a daily basis on the markets. It was based on the UK stock market’s volatility and its purpose was to protect the Swiss bank’s trading book from losses. Variance swaps quickly started to gain ground on Wall Street, especially in the late 1990s, when the Asian financial crisis hit and the collapse of hedge fund LTCM heavily shook the markets. This rose interest in the idea of trading volatility itself, but variance swaps remained too sophisticated for wide usage, at least until the beginning of the new century.
Despite the strong interest, the Vix has become a traded index only more recently: it summer 2002. A few years earlier, Mark Cuban had sold Broadcast.com, an internet radio company, to Yahoo for a cool $5.7bn, leaving him with $1.4bn worth of the portal’s stock. Then the dotcom bubble burst and Cuban wanted to buy some protection. He asked to a Goldman Sachs banker who tried to sell Cuban a variance swap, but the Texas-based billionaire wanted to get exposure to the “fear index” he had heard so much about. Soon afterwards he mentioned the businessman’s desire to punt on the Vix to a colleague, Sandy Rattray, which was stroke by the idea and immediately contacted the CBOE to make a proposal.Cboe hired the investment bank to develop Vix’s methodology and to make tradable futures contracts linked to it available on the market. In 2004, the first Vix futures was there. At the very beginning, interest was restricted to a few group of investors. But when the financial crisis erupted, the large demand for insurance on the market made the Vix index vary popular in the finance industry. People rushed to buy Vix futures to protect themselves against the heavily bearish market, and by 2008 more than 4,300 contracts were traded on a daily basis.
However, this is not all the story. Financial engineers soon opened the Vix market also to retail investors by constructing exchange-traded products based on the index. In 2009, Barclays built the first volatility-linked ETP using Vix futures, and by 2017 there were more than 40 Vix-linked ETPs available to trade. Among them, some are “long volatility” and some are “short” with an average daily trading volumes around 294,000 contracts last year. But, as February’s turmoil made clear, it very risky to have a relevant input into risk-management models that is also a popular trading tool.
Then, we get to the last February.
Launched in 2010, the Global Volatility Summit, an annual meeting of hedge funds dedicated to trading turmoil, has never been as largely and intensively attended as it was this year. The money manager pointed out the worrying implications of just $3bn of Vix-linked ETNs causing such market turmoil, highlighting how volatility had transformed from a proxy of risk into an input for risk, comparing it to the tail-eating uroboros snake of Ancient Greece mythology. During such event, Christopher Cole, founder of Artemis Capital Partners, warned that February’s events turbulence were just an appetiser for a “volatility revolution” that “will not be televised”. Cole alleged that there is now a “dangerous feedback loop linking volatility, low interest rates and financial engineering, and estimates that there is more than $2tn in strategies that both exert influence over, and are influenced by, stock market volatility”, a journalist at FT reported. “This has made volatility the only asset class that really matters”. Although XIV was of small matter, compared to the whole market it ended up being the spring that triggered a panic attack. Indeed, banks and investors hit by its collapse, strove to hedge their positions, in turn triggering a vast amount of automated selling by the volatility-targeting funds. Luckily, the pretty favourable economic environment smoothed the impact of such chaos, encouraging many investors to take advantage of the sell-off and gain from price bouncing. Cole added that the economy can’t be so buoyant every time such a shock happens and that the “XIV-triggered vol-mageddon was therefore just an amuse-bouche for what the future might hold”. “I don’t know if I’m crazy, or if the rest of the world is crazy,” he added. “But this is the kind of thing our children might look back and say, how did we not see this coming?” None has the right answer to such complicated questions. Nonetheless, people who have played a part in the history of volatility defend their role in its evolution while expressing concerns at the next stage. Also, volatility is now embedded in risk-management models. Trading strategies have been built on volatility targets. Making volatility a tradable asset has been, on the one hand, a valuable development, but, on the other hand, instruments such as the Vix ETNs proved to be very harmful. In a few years, our children might look back and say, how did we not see this coming?
Christopher Cole Physicists have long noted that observing some phenomena actually changes their nature. In finance a similar rule applies. As more people weave volatility into their models, they change its nature in subtle but important ways. But the point is: can or even should anything be done about it? Also, is it really so that the volatility-trading industry is big enough to shake the $80tn global stock market? Despite all the analysis and questions about what should or should not be done, Markowitz, the founder of volatility as proxy of risk, is sceptical that anything will ever change. “It has organised so much of our world that there’s simply no way of backing away from it,” he alleged.
Mattia Migliardi