Evolved volatility arbitrage strategy performs well in most market environments

Strategy evolution

twists-and-turns-of-a-yellow-rollercoaster

While some investors used volatility arbitrage for hedging purposes, now they are differentiating between volatility arbitrage and tail risk hedging. Vol arb should do well in most market environments.

Volatility arbitrage, much like the phenomenon on which it is based, has seen some significant ups and downs since 2008. The crisis proved to be a catalyst for investors’ growing interest in volatility. The vacillations of markets and subsequent returns as the crisis continues to unfold have sparked an evolution in the strategy.

Volatility arbitrage has become an increasingly diverse pool of funds, making it difficult for investors to get a holistic view and understand how the different funds fit into their portfolios. However, with volatility expected to remain volatile, interest in this asset class should continue to grow.

“The VIX was relatively unknown prior to 2008,” says Gilbert Keskin (pictured), co-head of volatility and convertible bonds at Amundi, which runs €5.9 billion ($7.9 billion) in volatility funds. “Now everyone is monitoring the VIX and interest in the asset class has increased markedly.”gilbert-keskin-amundi

Between 2004 and the end of 2007, markets were largely benign. The VIX, widely known as the ‘fear index’ and a key benchmark of expectations of near-term volatility measured by S&P 500 stock index option prices, averaged 14.6% over the period, well below the 20%-30% bracket considered ‘normal’ today.

Then came Lehman Brothers, which revealed the true extent of the financial crisis. Between August 29 and October 27, 2008, the VIX shot up from 20.68% to over 80%.

This proved a seminal moment in alerting investors to the need to consider the impact of volatility on portfolios. What followed was a dramatic increase in hedging tail risk through long volatility exposure. By making enormous profits when other parts of the portfolio were suffering, it offered a source of uncorrelated returns and some harbour from future storms.

But volatility is exactly that, volatile. By 2009 volatility had returned to more normal levels with the VIX averaging 31% during the year. “2009 and 2010 were very difficult years for long-only strategies and investors realised it costs a lot to be systematically long volatility in the medium term,” according to Keskin.

Time-decay effect
Being long volatility will inherently lose money over time, due to the need to roll futures as the volatility curve is in contango, so investors have to buy expensive and sell cheaply because of the time-decay effect of implied volatility.

Between February 27, 2009 and February 22, 2012, the iPath S&P 500 VIX Short-Term Futures ETN declined 94% from 454.84 to 25.81. Importantly, this does not constitute product under­performance but highlights the extent to which being long volatility is a loss-making strategy in the medium term.

volarb1-0312

The volatility arbitrage strategy saw outflows of $491 million and $2.6 billion in 2010 and 2009 respectively after posting the only negative performance for each year within the relative value bucket, according to Hedge Fund Research data. While the HFRX Relative Value Arbitrage Index was up 38.47% in 2009 and 7.65% in 2010, the HFRX Volatility Index (HFRXVOL) was down 3.19% and 4.36% respectively.

The Volatility Trading Index from prime broker Newedge, which tracks performance of an equally weighted portfolio of volatility trading and arbitrage funds including names such as Parallax, JP Capital, Amundi and BlueMountain, returned a negative 2.07% in 2009 and was down 2% in 2010.

As investors realised they would continue to make losses as long as volatility and the volatility of volatility remained low, it sparked a growing interest in benefiting from volatility more cost efficiently by getting exposure to the potential upside without having to accept this as a loss-making strategy.

Investors moved into more complex volatility exposures that benefit from both upwards and downwards movements in volatility, which is mean reverting. The result was a realisation that volatility can be traded as a standalone asset class.

“Historically some investors looked at volatility arbitrage for hedging purposes,” says Vinod Kanbi (pictured), senior strategy analyst at Axa Investment Managers. “Now they are increasingly differentiating between volatility arbitrage and tail risk hedging. Volatility arbitrage is not just about creating portfolios that protect investors when volatility increases. Some do the opposite. It is an absolute return strategy in its own right,” he explains.vinod-kanbi-axa-im

The result was a proliferation in systematic, directional volatility strategies that seek to exploit the difference between implied volatility, which is an estimate of future volatility, and realised volatility.

Generally, the spread is 4% on average for one-month S&P 500 index implied volatility versus one-month realised volatility, says Kanbi. “However, this is an average number with deviation around this mean. It also does not hold as a rule across all strikes and maturities and can vary widely.”

Volatility arbitrage managers can harvest that premium by buying or selling options while hedging equity and interest rate risks.

Being short volatility will be painful when volatility spikes because realised volatility will be higher than implied volatility. However, these funds will generally be betting the other way preceding a volatility event.

“You would think when volatility dominates, the merits of the strategy would come to the fore but that has not been our experience,” notes a senior fund of hedge funds manager.

Volatility arbitrage strategies as measured by the HFRXVOL have tended to make a sharp downward move when volatility spikes. This was true in late 2008 following the collapse of Lehman Brothers, in 2010 when the flash crash hit and in 2011 as the eurozone crisis unfolded.

In each of those periods volatility hedge funds based on the HFRXVOL fell dramatically. The loss faced by those that are short volatility when it spikes will increase as the difference between realised and implied volatility increases. So the sharper the spike, the greater the loss.

Realised volatility has been significantly higher than implied volatility each time there has been a spike in volatility.

According to Udi Sela, vice president at independent analytics provider Numerix, most hedge funds using volatility arbitrage strategies are directional in nature and are short volatility. “This was particularly the case prior to 2011,” he says. “Funds did not correctly estimate the cost of being tied into this position.”

In the months leading up to both the collapse of Lehman Brothers and the summer of 2011, markets had been expecting volatility to remain benign over the short term. The daily value for the VIX was 23% on average for the year to August 28, 2008. For the three months from March 23 to June 14, 2011, the daily VIX level remained below 20%, averaging 17%, and hitting 14.62% on April 28, the lowest level since June 2007.

volarb2-0312

Volatility events can also hurt those that are systematically short volatility in the medium term. The average daily premium between the VIX and one-year implied volatility on the S&P 500 was down 1.4% between November 20, 2008 and January 31, 2012 and fell 1.2% between June 30, 2000 and January 31, 2012.

The losses suffered from being short volatility, and the realisation such exposure would not offer ­decorrelated returns when they were most needed, seems to have forced another rethink among investors. Meanwhile, the strategy itself has evolved into an increasingly complex pool of funds using a wide variety of ways to trade volatility in different forms.

“Even within volatility, when people talk about the strategy it means many things to many people. Volatility arbitrage is a broadly defined space that has negative connotations as many of the previous highfliers no longer exist or grew too quickly,” says Stephen Yashar, co-founder of Cosyne Capital Management, which focuses on enhanced relative value volatility strategies.

Divided strategies
Hedge funds by their very nature dislike being classified, but the strategy appears to have divided into systematic and more active strategies and, within that, into those that bet on the direction of volatility in one form or another, whether that be long, short or long/short, and those that focus on ‘pure’ volatility arbitrage through relative value strategies.

Today’s volatility funds, whether they belong to the pure arbitrage/relative value space or focus on the direction (or term structure) of volatility have a similar goal in common: to provide a decorrelated source of return that makes money in any environment.

As a result managers are more likely to be market neutral, relying less on which way volatility is moving and more on the volatility of volatility to increase dislocations and subsequently their opportunity set.

“We do pure relative value volatility arbitrage, which means we try to have little sensitivity to levels of volatility. This should afford us a return profile that is uncorrelated to the S&P 500, VIX or any other major index, offering investors a way to diversify returns,” explains Nelson Saiers (pictured), chief investment officer of volatility relative value manager Alphabet Management.nelson-saiers-alphabet

This smarter approach to volatility helped push the strategy into positive territory in 2011 despite spikes in volatility in the second half of the year. During 2011 the HFRXVOL returned 1.42% while the broader HFRX Relative Value Arbitrage Index fell 4%.

The Newedge Volatility Trading Index also showed volatility hedge funds were better able to leverage the greater volatility of volatility experienced in 2011. The index returned a negative 2.07% in 2009, minus 2% in 2010 and was up 1.32% in 2011.

“The fact the Newedge Volatility Trading Index outperformed the average hedge fund in 2011 as well as the S&P 500, albeit slightly, is evidence the strategy had fared pretty well in the recent past,” says Saiers.

Volatility arbitrage funds’ performance also shows they are less volatile than the VIX, effectively generating returns from, but also protecting investors from, the volatility of volatility. As of the end of January 2012, Amundi’s €1.75 billion ($2.35 billion) Absolute Volatility Euro Equities fund, for example, had shown volatility of 8.72% since inception in November 2006 versus volatility of volatility of 47.58%. The fund has returned 60.17% since inception.

“Directional volatility funds will be more volatile than pure arbitrage strategies, but the expected returns will also be higher. Our arbitrage funds have an expected return of cash plus 2% where our directional volatility funds expect to generate around 7% per annum over a three-year horizon,” says Amundi’s Keskin.

The attractions of accessing the volatility risk premium has now become so widespread the average daily volume of VIX options increased 58% between 2010 and 2011 to 391,993 and was already 5% higher by February 17, 2012 at 413,052. This has increased overall liquidity in volatility securities.

“The proliferation of ETFs and their associated options has increased the scope of asset classes available to ‘equity-based’ volatility arbitrage managers. The greater liquidity and transparency typically seen with a large number of these listed instruments has forged a lot of opportunities as it allows us to trade a myriad of products and take advantage of more mispricings,” notes Alphabet Management’s Saiers.

This should prove particularly bene­ficial to funds trading volatility securities on indexes. However, for those interested in single names, the liquidity pool has moved in the opposite direction, a trend that could prove disastrous for some relative value strategies.

“Lower single-name liquidity has been a continued trend since 2008. The stock-specific risk is too high and I don’t see liquidity returning for the next five years, even if the market returns to fundamentals sooner, and that will only happen if things calm down and the market forgets. That means there is not sufficient liquidity for dispersion trades, for example,” says Nadejda Rakovska, head of business development at €30 million ($40.3 million) volatility arbitrage fund Volvar.

Dispersion strategies, which bet on the difference between the volatility of an index and its constituent parts and therefore requires holding positions in a large number of single-name volatility securities, have historically accounted for a significant portion of volatility arbitrage funds, helping to explain why the HFRXVOL index falls as volatility spikes.

“A lot of dispersion strategies were hurt during the last quarter of 2011, especially as OTC liquidity dried up,” says Yashar, who does not run a dispersion strategy. “Many were a proxy for being short volatility. When things get difficult is exactly when they need liquidity to adjust their positions, but the OTC market is not deep enough to accommodate them when they are all running for the door.”

In fact liquidity in single names has become so tight that one trader suggests there were only a few hundred with a liquidity profile that made it worth trading them in any market environment.

Short-date bias
In general liquidity is also biased towards short-dated securities. “Investors usually use derivatives on short-term maturities and around two-thirds of options volumes are for maturities of less than a year,” says Amundi’s Keskin.

Using short-dated maturities can be an advantage to option strategies, since they provide an extra layer of protection when markets are acting irrationally. “Keynes famously said that markets can remain irrational longer than investors can remain solvent,” Alphabet’s Saiers says.

“The short-dated nature of these securities ensures that a manager is much less beholden to the markets’ whim as the trade will mature in the coming days or months. In other strategies such as long/short equity or relative value credit whose maturities are either indefinite or long-dated, the manager may not have the staying power to profit from a good investment thesis,” Saiers adds.

volarb3-0312

However, with shorter maturity comes greater cost because of futures rolling and time decay (the loss of value on options as time passes). Both time decay and rolling costs are more significant on short-term positions because the futures have to be rolled more frequently and the impact of each passing day on the value of the option in question is greater than for longer-dated securities.

Amundi focuses on one-year implied volatility. “This gives us the advantage of lower roll costs and time-decay, but there is still enough liquidity at this level,” Keskin says. “This is something investors need to consider as short maturity securities mean greater cost even if it is more sensitive to spikes in volatility, making market timing much more important.”

Longer-dated maturities are less sensitive to short-term non-structural shocks and are therefore a better representation of structural changes in volatility. This is arguably a more measured approach for those exploiting the difference between implied and realised volatility as the number of spikes where realised volatility exceeds implied volatility will decrease.

With headline risks likely to continue dominating markets, the way the eurozone crisis is handled will be a big driver of volatility over 2012. Many experts predict a continued ‘muddle through’ approach. But with the prospect of political change in the US, France and China, investors should prepare for the continued volatility of volatility.

“However, it is very important investors understand what volatility arbitrage is and how it fits into portfolios,” notes Cosyne Capital Management’s Yashar. “Volatility arbitrage is a source of absolute return and should do well in most market environments but outperform in high volatility of volatility periods. It therefore does well when the drivers of volatility are hurting other areas of investors’ portfolios so it is a good source of uncorrelated returns.”

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@risk.net or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact info@risk.net to find out more.

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here