We are experiencing some of the largest price swings in the history of the stock market, with hundred-point five-minute bars in the Dow Jones Industrial Average common. To survive in this market, you must learn how this volatility can affect your trading strategies. You must learn how to adapt. To that end, we will discuss ways to measure volatility and examine how volatility risk affects mechanical trading strategies on a risk-adjusted basis.
We have seen several days in the past month where the Dow has had a 600-plus-point range. Before the current mortgage crisis, this level of relative volatility occurred once or twice every five years. To understand how to best operate in this environment, it’s important to examine what created it.
The subprime mortgage financial crisis is blamed for losses in the billions of dollars and forcing many financial institutions and individuals into bankruptcy. It has pushed the entire economy to the brink of a crisis that could, if unchecked, lead to bank failures on par with those of the Great Depression.
The present crisis has many causes: the lack of checks within the mortgage industry itself, rising oil prices, high food and transport prices and inflationary pressures. The economy slowed. Mortgages were defaulted on and recessionary pressures increased. It all began, however, with the mortgages. In general, there are two types of mortgage loans offered in the United States: fixed-rate mortgages, where the interest rate is fixed for a set term, usually 15 or 30 years and adjustable-rate mortgages (ARM), where the interest rate is variable, often following a period (generally one to seven years) of a fixed rate.
The roots of the subprime crisis are planted deep in this country’s previous financial meltdown, the bursting of the Internet bubble around 2002. After the dot-com bubble burst, the Federal Reserve cut their target for the Fed funds rate from 6.5% to 1%. This resulted in a massive easing of credit. Banks were able to lower their requirements on personal loans. Interest rates on 30-year fixed-rate mortgages fell by 2.5% to a historical low of about 5.5%, on average. The interest rate on one-year adjustable-rate mortgage rates hit 4%.
These loans were one of the reasons house prices increased so quickly. The payment amounts, not the loan amounts, were the price people saw for a house. As a result, new buyers flooded the market and many were eager to invest in rising real estate assets. Then, in 2006-07 when interest rates rose and re-financing became more difficult, defaults and foreclosures increased rapidly and home prices stagnated. The pain quickly spread into the commercial side, which had become just as inebriated on the profits in commercial paper representing all these bad mortgages.
Ironically, this pain was potentially exacerbated by three government-sponsored enterprises (GSE) designed to help more people buy a home. These include:
1. Fannie Mae (The Federal National Mortgage Association), a stockholder-owned corporation chartered by Congress in 1968 as a GSE, but founded in 1938 during the Great Depression.
2. Ginnie Mae (Government National Mortgage Association) became the guarantor of government-issued mortgages in 1968.
3. Freddie Mac (The Federal Home Loan Mortgage Corporation), founded to compete with Fannie Mae in 1970.
GSEs GET INVOLVED
GSEs buy large numbers of mortgages from banks and issue mortgage-backed bonds or mortgage-backed securities (MBS) at lower interest rates to investors. This created more mortgages and the public used these mortgages to purchase homes. Thanks to new technology, this process grew in efficiency and allowed for faster and easier outflow of capital from investors to borrowers.
Automated underwriting was the new thing. This included Freddie Mac’s Loan Prospector and Fannie Mae’s Desktop Underwriter software programs. Not only were loan requirements minimal, but mortgage companies were able to buy, sell and structure loans with a few keystrokes. Low-rate ARMs, interest-only loans and balloon-payment loans allowed homebuyers to qualify for purchases that they never could before — that is, until the grace period on the low rates or the balloon payment came due.
The system is a good one, with one big stumbling block. There is no safety net in place for a major drop in housing prices. When prices drop 20% in a year, buyers suddenly owe more on their houses than the homes are worth. Consider a hypothetical buyer who had purchased a $300,000 house with almost no money down a year earlier. If that home is now worth $240,000, or less, just a year later, he may decide the nearly $60,000 difference is well worth ruining his good credit.
Another piece of the mortgage crisis puzzle is the disintermediation of investor funds, which gravitated from traditional savings accounts to direct market instruments, such as money market funds. The capital reserve requirements of the investment banking arms selling these instruments are more lenient than the requirements of depository banks. Some investment banks lend as much as 30 times the value of their capital.
This is incredible leverage, and when investors panicked and withdrew money from the money market funds, it resulted in margin calls on the investment banks and mortgage companies, and these firms incurred huge losses, became bankrupt or merged with others. The pain has been quick and ruthless for homebuyers, as well. In March 2007, the value of U.S. subprime mortgages was more than $1 trillion, with more than $7 million of which being first-lien, subprime mortgages. In October 2007, failed subprime ARMs were almost three times that of 2005. Foreclosures rose rapidly by almost 80% from 2006 to 2007.
The liquidity concerns, homebuyer suffering and prospects for worsening conditions encouraged central banks globally to provide funds to member banks to encourage lending to worthy borrowers and restore faith in the commercial paper markets. In the United States and United Kingdom, bailouts ran rampant (see “Government bailouts”).
An obvious result of this massive financial stress and uncertainty has been volatility across all markets, particularly the equity markets.
In the equity markets, however, periods of high volatility are nothing new. Previous hyper volatile periods include 1929, 1987, 1998, 2002 and 2008. Consider the volatility in the Dow from 1933 to September 2008. “Historical perspective” uses monthly percentage changes to calculate a normalized standard deviation.
We can define high volatility as a 12-month standard deviation above 5%, which has occurred many times. We had five events with peaks about 6%: during the consequences of the Great Depression, during the summer of 1938 (German aggression before World War II), during the great inflation of 1975 and near the end of the early 1970s bear market, the crash of 1987, and during the great 2002 bear market. These peaks often lasted several months.
ENTER THE VIX
The Volatility Index, or VIX, is among the widely accepted methods to measure stock market volatility. The Chicago Board Options Exchange (CBOE) developed the first version in 1993; it was calculated by taking the weighted average of implied volatility for the S&P 100 index (OEX) calls and puts. The next revision in September 2003 gave a more accurate depiction of broad market volatility by using options on the S&P 500. The VIX is often referred to as the fear gauge.
The VIX doesn’t measure the volatility of a single issue or option instrument, but uses a wide range of strike prices of various calls and puts. This result is a more accurate measure of the market’s expectation of near-term volatility. The VIX calculation is independent of any theoretical pricing model, using a formula that averages the weighted prices of at-the-money and out-of-the money puts and calls to derive expected volatility.
Although the VIX is often called the fear index, a high VIX isn’t necessarily bearish for stocks. Instead, the VIX is a measure of fear of volatility in either direction, including to the upside. High VIX readings mean investors see significant risk that the market will move sharply, whether downward or upward. The highest VIX readings occur when investors anticipate huge moves in either direction. The VIX will be low only when investors perceive neither significant downside risk nor significant upside potential.
Since its revision, the importance of the VIX has grown. Now we have both VIX futures and options. The futures started trading in March 2004 at CBOE, while the options started trading in February 2006.
During the current crisis, the VIX has set three new record highs in October 2008: on Oct. 9, it hit an intraday high of 64.92; on Oct. 10, it hit an intraday high of 75.92; on Oct. 24, it had an intraday high of 89.53.
The VIX indicator incorporates a cross-section of investor sentiment. It has an inverse relationship to the market. Low VIX values, 20-25, occur when the market is somewhat uninterested, especially during slow grinding rallies, and often will lead to a market selloff. VIX is often used as a contrarian’s indicator. Prolonged or extremely low VIX readings indicate a high degree of complacency. Some contrarians view readings below 20 as excessively bearish, though the bull market following the 2002 equity low occurred with the VIX below 20.
Prolonged or extremely high VIX readings indicate a high degree of anxiety among options traders (bullish). High VIX readings usually occur after an extended/sharp decline and sentiment is still quite bearish. Some contrarians view readings above 30 as bullish. Conflicting signals between VIX and the market also can provide sentiment clues for the short term. Overly bullish sentiment (complacency) is considered bearish by contrarians, while overly bearish sentiment (panic) is considered bullish.
If the market declines sharply and VIX remains unchanged or decreases in value (toward complacency), it could mean that decline will continue. Contrarians might feel that there isn’t enough bearishness or panic in the market to warrant a bottom. If the market advances sharply and the VIX increases in value (toward panic), it could mean that the advance will continue. Contrarians might feel that there isn’t enough bullishness or complacency to warrant a top.
Looking at a chart of the VIX, we can see that we’ve easily surpassed levels reached during the implosion of hedge fund Long-Term Capital Management, 9/11 and the bottom of the bear market in 2002 (see “How high?”).
VOLATILITY & PRICE
Volatility’s relationship with price is complex. In short, some markets are positively correlated to volatility, while others are negatively correlated. Consider these two examples, Treasury bonds and crude oil. Both of these markets are negatively correlated to volatility (low volatility is bullish, while high is bearish). We trade a divergence of this relationship. Let’s take a closer look at crude oil.
In crude oil, when volatility drops too low (bullish), a major rally in crude occurs. Crude is unlike equity markets where volatility is higher during bear markets. In crude, volatility increases during bull markets until it gets so high that people become afraid to stay long and liquidate these positions and cause major sell-offs.
We created a system that takes advantage of this concept. First, we define volatility as range, a combination of both short-term and long-term range. Hence, we multiply average range and range. We then normalize range over a longer period so we can gauge how current volatility relates to past values. If crude is falling and volatility goes below a given level, we buy. If crude is rising and volatility goes above a given level, we sell. (You can see the code for this system at www.futuresmag.com under downloads.)
Tested over the period Jan. 4, 1983, to Oct. 17, 2008, with the parameters 20, 90, 16, 0.5 we get the following results:
Total net profit $141,820.00
Open position P/L -$59,200.00
Number of trades 289
Percent profitable 61.59%
Largest winning trade $20,750.00
Largest losing trade $17,620.00
Average winning trade$2,108.20
Average losing trade$2,141.65
Max intraday drawdown$62,400.00
Profit factor 1.61
Yearly return on account 9.17%
Account size required $62,400.00
We made $127,350 on the long side and $14,470 on the short side. We won almost 68% of the trades on the long side and almost 56% on the short side. The idea here isn’t to present a tradable system — no deductions were taken for slippage or commissions — but to show that volatility can be predictive of market movement. Its current open losing trade was a buy signal recently. When volatility fell, as prices rose, it went long at 131 and got caught.
EFFECTS ON TRADING PLAN
When developing a money management program, the bigger the risks you take, the higher your returns. This is true in general, but when risks get too high, it doesn’t always mean a similar increase in profit.
We can explain this concept with two versions of a channel breakout. One takes all the trades on a given portfolio, and the other skips trades if the risk on a given trade is above a given level. The code for both of these systems also can be found at www.futuresmag.com under downloads.
We shall test on a basket of 25 markets that includes the major currencies, agricultural commodities, metals, energies, meats and financials.
We’ll use a standard 20-bar breakout for both the entries and exits (a classic length). Using this allows us to concentrate on limiting risk. We tested combinations allowing risk of $1 million. Using such a large number effectively allows us to turn this filter off.
On a risk-adjusted basis, $2,500 risk per trade produces the best returns. Our results at the level are as follows, with $100 deducted for slippage and commissions:
Total net profit $803,556.54
Open position P/L$32,767.20
Number of trades 3,420
Percent profitable 37.75%
Largest winning trade$41,262.50
Largest losing trade $4,262.50
Average winning trade $2,302.19
Average losing trade-$1,019.54
Max intraday drawdown -$66,464.74
Profit factor 1.37
Yearly return on account 42.00%
Account size required$66,464.74
When we limit risk to $2,500, our profit-to-drawdown ratio is more than 12-to-1 vs. 8.45-to-1 if we don’t limit it. More risk does not always equal more reward, at least when time constraints fail to give the additional risk the time to work.
The area of high volatility is ripe for additional study. With the markets now at historic peaks, it’s a critical time to put volatility analysis at the forefront of your market studies. Research shows, however, that you need to examine volatility in a market-specific context. What volatility suggests for one may not be relevant to another.
One piece of advice is universal, however. Work harder and be more prepared for swift, sharp moves in either direction. You may not always be able to take volatility to the bank, but high volatility can almost always take you to the cleaners.
Murray A. Ruggiero Jr. is a consultant. His firm, Ruggiero Associates, develops market timing systems. He is the author of "Cybernetic Trading Strategies" (Wiley). E-mail him at firstname.lastname@example.org.