RUSSELL INDEXES RESEARCH
RVX futures as a hedge
for small cap–specific risk
Barry Feldman, Ph.D., CFA
EXECUTIVE SUMMARY:
1. Volatility products can be used to hedge downside risk in equity portfolios
2. RVXSM futures provides a U.S. small cap equity focused alternative to
CBOE Volatility Index® (VIX® Index) futures
3. RVX futures hedges can potentially be more effective when the risk is
specific to small cap stocks
Summary
RVX volatility futures (CBOE Russell 2000® Volatility Index Futures Contracts) track the
implied volatility expectations of the U.S. small cap equity market and can be a particularly
effective hedge for the risk inherent in the small cap asset class. The potential usefulness of
RVX futures in hedging small cap–specific risk is illustrated in this paper with results of a
simulation case study that includes two hedging examples based on market movements in
spring and summer 2014. One example shows that a simulated small cap portfolio could have
been better hedged with RVX futures than with VIX futures during the period examined.
This paper starts with a brief introduction to volatility products and the concept of small cap–
specific risk. It then presents the case study results, starting with a review of market
conditions leading up to the small cap market correction in 2014 and continuing through to a
comparative estimation of the results of a particular hedging strategy implemented with both
RVX and VIX volatility futures. RVX trades are conservatively modeled. A methodology
appendix provides details of how the hedges are constructed and valued. It also includes an
examination of the differences between volatility indexes and the prices of volatility futures
based on those indexes, basic details of the construction of the futures hedges, and the
approach used here to estimate the prices used in implementing and rolling RVX and VIX
hedges.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
JUNE 2015
Introduction
VIX-based volatility products have expanded over the last 10 years – from futures contracts,
to volatility options, and to increasingly sophisticated ETFs (e.g., VXX and XIV). VIX, the
CBOE Volatility Index, is a measure of the implied volatility of the U.S. large cap asset class
as represented by the S&P 500®.1 The implied volatility of a security is determined by the
prices of options on the security and is an estimate of the market-implied expected volatility
over the remaining life of the options.2
RVX is the implied volatility index for the U.S. small cap asset class as represented by the
Russell 2000 Index.3 Small cap–specific risk, which is driven by factors associated with small
cap stocks, is a type of asset class–specific risk. Many risk factors affect all asset classes.
The asset class–specific risk examined in this case study is the market perception that small
cap stocks might be overpriced relative to the market as a whole. The greater liquidity of the
VIX contract has made it a frequent choice for managers seeking to hedge small cap stocks
against market-wide risks. Yet the examples developed here show that hedging with VIX may
not be the best strategy for small cap–specific risks.
Our baseline assumption is that an investor wants to use volatility futures to hedge a small cap
portfolio. (Options offer another tool for hedging portfolio risk, but will not be studied here.4)
VIX positions are modeled with entry and exit trading at the contract daily settlement price.
RVX entry/exit trades are modeled more conservatively by taking the observed bid/ask spread
into account, with purchases at the daily settlement price plus one-half the daily quote spread
and sales at the daily settlement price minus one-half the daily quote spread. Thus simulated
RVX purchases are at the ask price, and simulated sales are at the bid price. The primary
cost of maintaining a volatility futures hedge is typically the cost of rolling the contracts, which
is when contracts nearing expiration are sold and replaced with longer-dated contracts. VIX
contracts are rolled at average time-matched trade prices for both contracts. RVX contracts
are rolled conservatively, based on the average time-matched bid price for the expiring
contract and ask price for the next-month contract. The methodology appendix provides
further details.
1
VIX information may be found at cfe.cboe.com/Products/Products_VIX.aspx. CBOE, Chicago Board Options Exchange, CFE, CBOE Volatility
Index, and VIX are registered trademarks of Chicago Board Options Exchange, Incorporated (CBOE). RVX is a service mark of CBOE. The Russell
2000 Index is a registered trademark of FTSE Russell Indexes, used under license. CBOE data is believed to be correct but CBOE does not
guarantee the accuracy of the data and will not be held liable for consequences of its use.
2
During the period of this study, both VIX implied volatility and RVX implied volatility are based on the prices of the two nearest-to-expiration
monthly options contracts. Contracts at all strike prices are included in the calculation. The prices of these contracts are weighted so as to create a
constant 30-day forward expected volatility horizon. Starting fourth quarter 2014, the VIX methodology changed; it now uses the nearest expiring
weekly SPX options.
3
RVX futures were recently relaunched. RVX trading was discontinued in the spring of 2009 and reintroduced in the fall of 2013. RVX trading still
displays limited depth and liquidity. During 2009, RVX trading dwindled, due to a combination of factors that included the financial crisis of the time
and technical issues around contract expiration. CBOE information on RVX futures can be found at cfe.cboe.com/Products/Products_VU.aspx.
Further information about RVX volatility futures contracts can be found in “Russell 2000 Volatility Futures,” Barry Feldman, Russell Research,
October 2013.
4
The interested reader can find a comparison of the relative merits of index options, volatility futures, variance swaps and options on volatility
futures in “The Art of Hedging,” QDS Vega Times, Morgan Stanley, June 14, 2014.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
2
Hedging the 2014 small cap correction
The spring of 2014 was a time of increasing investor uncertainty about U.S. small cap equity
prices. Many believed small cap was overvalued. On February 6, 2014, for example, a
Bloomberg article noted that implied volatility on iShares Russell 2000 ETF options had
spiked and that strategists were seeing a downturn coming for small cap prices.5 This
uncertainly was reflected in RVX values. Figure 1 shows the RVX Index from July 1, 2013, to
February 6, 2014. A spike in small cap implied volatility can plainly be seen to have started on
January 24, 2014. A peak can be seen on February 3, followed by a small decline to February
6, as highlighted in red.
Figure 1 / RVX Index values
July 1, 2013 to February 6, 2014
30
25
RVX
20
15
10
5
1/27/2014
1/13/2014
12/30/2013
12/16/2013
12/2/2013
11/18/2013
11/4/2013
10/21/2013
10/7/2013
9/23/2013
9/9/2013
8/26/2013
8/12/2013
7/29/2013
7/15/2013
7/1/2013
0
Source: CBOE. February 6, 2014 highlighted.
Figure 2 displays the RVX premium relative to VIX. The RVX premium is defined as RVX
Index values as a percentage relative to VIX Index values. A zero premium corresponds to
RVX and VIX indexes having the same value. The RVX premium is a representation of the
market perception of the magnitude of small cap risk relative to large cap risk. An increase in
expected small cap–specific risk should be reflected in an increase in the RVX premium.
Looking at the period January 1, 2014 through August 31, 2014, it can be seen in Figure 2
that there is also a spike in the RVX premium starting on January 24. The RVX premium value
of 40.9% on February 6 is highlighted in red, both to provide orientation for the different time
periods shown in Figures 1 and 2 and to show that the RVX premium continues to spike on
this date, even though the RVX values in Figure 1 decline. This indicates that VIX values
declined faster than RVX values from February 3 to February 6. Figure 2 displays the RVX
premium until August 31, 2014. After February 6, the premium first reverts back into the range
in which it had been trending and then trends unevenly upward through at least the beginning
of June. One interpretation of this upward trend is that full risk of an expected small cap
correction was not immediately reflected in options prices and volatility indexes as of February
6. It can also be seen that the RVX premium declines sharply at the end of July, suggesting a
potential end to the period of small cap–specific risk.
5
See “Small-Cap Bears at Highest Since ’12 Selloff: Options,” February 6, 2014, at Bloomberg Online, bloomberg.com/news/2014-02-06/smallcap-bears-at-highest-since-12-on-selloff-options.html.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
3
Figure 2 / RVX premium, defined as RVX/VIX-1
January 1, 2014 to August 31, 2014
80%
70%
RVX Premium
60%
50%
40%
30%
20%
10%
8/28/2014
8/14/2014
7/31/2014
7/17/2014
7/3/2014
6/19/2014
6/5/2014
5/22/2014
5/8/2014
4/24/2014
4/10/2014
3/27/2014
3/13/2014
2/27/2014
2/13/2014
1/30/2014
1/16/2014
1/2/2014
0%
Source: CBOE. February 6, 2014 highlighted.
The simulation case study is based on the time period January 1, 2014 to August 31, 2014,
depicted in Figure 2. This is obviously a retrospective study. The selection of the times to
hedge was based on the benefit of a hindsight-based timing ability that is not available to
investors. The dates that define the first hedge were selected according to the idea that they
might have been chosen by an investor with moderate timing ability. So, for example, the
small cap volatility spike starting on January 24, which has already been discussed, is not
used as a hedging example. It seems unlikely that an investor with moderate timing ability
would have been able to predict this volatility spike with sufficient accuracy.6
The first hedge is implemented on February 24, about two weeks after the end of the volatility
spike and publication of the Bloomberg article. This gives our hypothetical investor time after
the February 6 spike to do an assessment of the small cap market and to design and
implement a hedge.
The second hedge is started on June 30 and ends on August 6. The primary purpose of this
second hedge is to compare RVX and VIX hedge performance in a changing market
environment, wherein small cap–specific risk is not as prominent.
The dates for ending the hedges are not crucial. A change of a few days does not affect the
qualitative results.
Figure 3 shows the effects of hedging Russell 2000 Index performance over the two periods
just described. The cumulative total return of the Russell 2000 is shown, along with
cumulative performance when the Russell 2000 is hedged using RVX futures or VIX futures.
The shaded areas indicate the time periods of the hedge examples. The RVX-based hedge
outperformed the Russell 2000 and the VIX-based hedge strategy. The VIX-based hedge was
initially successful, in the early period of the first hedge, but quickly had increasing periods of
underperformance relative to the Russell 2000. It is only toward the end of the second
hedging window that the VIX-based hedging strategy outperformed the Russell 2000 on a
6
However, an investor with superior timing ability could have entered a RVX or VIX volatility hedge at the January 22 settlement price and sold at
the February 5 settlement price to realize a return of 43.2% for an RVX position and 38.3% for a VIX position, not taking the minimal costs of
establishing and liquidating positions into account.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
4
sustained basis. As of August 31, 2014, the RVX-based hedge outperformed the Russell 2000
by 12.95% and outperformed the VIX-based hedge by 7.60%.
Figure 3 / Russell 2000 Index cumulative total return performance and the
performance of two Russell 2000 50%-hedged strategies, one using RVX and one
using VIX futures
First hedge period is February 24, 2014 to May 8, 2014. Second hedge period is June 30, 2014 to
August 6, 2014. Data is from Russell Indexes and CBOE. Details of hedge construction are
provided in this paper’s appendix. Complete time period is January 1, 2014 to August 31, 2014.
120
Hedge 1
January 1, 2014 = 100
115
Hedge 2
110
105
100
95
90
85
Russell 2000
8/28/2014
8/14/2014
7/31/2014
7/17/2014
7/3/2014
6/19/2014
6/5/2014
5/22/2014
5/8/2014
4/24/2014
4/10/2014
3/27/2014
3/13/2014
2/27/2014
2/13/2014
1/30/2014
1/16/2014
1/2/2014
80
Russell 2000 plus 50% RVX Hedge
Russell 2000 plus 50% VIX Hedge
Source: CBOE and FTSE/Russell Indexes.
Table 1 presents performance statistics for the complete study history and for the time periods
of the two hedges. The results over the complete study history show that the RVX hedge
protected investor value with minimal increase in portfolio volatility. The maximum drawdown
statistics for the complete history time period show that the RVX hedge reduced maximum
drawdown from 9.06% to 7.40% while the VIX hedge had a 17.00% maximum drawdown. The
cumulative return of the RVX hedged portfolio was also considerably better. Note that the
complete history statistics in Table 1 involve volatility hedges only over the two time periods of
the hedging examples.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
5
Table 1 / Summary performance statistics for the Russell 2000 and RVX-hedged
and VIX-hedged Russell 2000 strategies as defined here
Results presented for the entire study history and for the two hedge periods: February 24, 2014 to
May 8, 2014 and June 30, 2014 to August 6, 2014.
TIME PERIOD
STATISTIC
Complete history:
January 1 to
August 31, 2014
Average daily return
Cumulative return
RUSSELL 2000
RUSSELL 2000
PLUS 50% RVX
HEDGE
RUSSELL 2000
PLUS 50% VIX
HEDGE
0.02%
0.10%
0.07%
2.89%
16.21%
10.70%
Annualized standard
deviation
15.70%
16.55%
20.57%
Maximum drawdown
9.06%
7.40%
17.00%
0.06
-0.02
-0.45
-0.06
0.22
0.55
1.70
1.78
Information ratio
Daily skew
Daily kurtosis
Hedge 1:
February 24 to
May 8, 2014
Average daily return
-0.12%
0.00%
-0.17%
Cumulative return
-6.30%
-0.15%
-8.96%
Annualized standard
deviation
16.92%
15.65%
23.30%
Maximum drawdown
8.98%
7.18%
16.92%
--
0.08
-0.02
-0.12
0.08
0.55
Information ratio
Daily skew
Daily kurtosis
Hedge 2:
June 30 to
August 6, 2014
Average daily return
0.52
0.18
1.93
-0.24%
0.13%
0.23%
Cumulative return
-5.58%
3.42%
5.91%
Annualized standard
deviation
14.96%
22.34%
29.98%
Maximum drawdown
7.67%
6.10%
6.29%
--
0.19
0.20
Daily skew
-0.22
0.63
0.40
Daily kurtosis
-0.19
2.79
0.04
Information ratio
Source: Russell Indexes and CBOE.
Figure 3 shows the RVX functioned successfully as a hedge in the first period, almost
completely neutralizing the Russell 2000 Index loss in value. This result was achieved with
lower levels of volatility than experienced by the Russell 2000. VIX failed as a hedge in the
first time period. This failure is visible both in the low returns and the large drawdown reported
in Table 1.
The results for the second hedge reported in Table 1 show that the RVX hedge performed
effectively and provided positive net returns, but was outperformed by the VIX hedge. The
performance of the RVX hedge relative to the Russell 2000 index is 9.53%, and that of the
VIX hedge is 12.17%. Note, however, that the VIX hedge outperformance is accompanied by
considerably higher volatility. As a result of this increase in volatility, the information ratio of
the VIX hedge is only marginally higher than that of the RVX hedge.
The likely reason for the strong performance of the second VIX-based hedge can be seen in
Figure 4. This chart shows a long period of steady declines in VIX values that starts in April
and reverses over the month of July, before spiking on July 31 (highlighted in red). This spike
occurred on the same day that the Dow Jones industrial average fell 317 points, as a result of
Russell Indexes // RVX futures as a hedge for small cap–specific risk
6
investor fears relating to domestic and global concerns.7 The RVX premium declines sharply
at the time of the VIX spike and stabilizes for the first time since the February 6 spike, in a
range mostly below that spike. The second hedging period can thus be seen to embody a
transition from small cap stock risk being driven primarily by asset class–specific issues to
being driven by market-wide factors.
Figure 4 / RVX premium; RVX and VIX values
January 1, 2014 to August 31, 2014
30
70%
25
RVX premium
60%
20
50%
40%
15
30%
10
20%
5
10%
RVX Premium
VIX
8/28/2014
8/14/2014
7/31/2014
7/17/2014
7/3/2014
6/19/2014
6/5/2014
5/22/2014
5/8/2014
4/24/2014
4/10/2014
3/27/2014
3/13/2014
2/27/2014
2/13/2014
1/30/2014
1/16/2014
0
1/2/2014
0%
RVX and VIX Index values
80%
RVX
Source: CBOE. July 31, 2014 highlighted.
Conclusions
A volatility hedge is utilized to protect against the marketplace volatility under the assessment
that the likely benefits outweigh the more certain costs entailed. Because a hedge is costly to
maintain, some ability to time implementation and liquidation is important.
The effect of the lower liquidity of RVX is modeled in this study by taking positions at ask
prices and liquidating positions at bid prices. VIX contracts, in contrast, are modeled to trade
at average and settlement prices. These adjustments provide for a more conservative
estimate of the performance of RVX hedges relative to VIX hedges.
One result of the conservative pricing approach is that the RVX continuous hedge
underperformed the VIX continuous hedge over the total study period. In spite of this
handicap, the RVX hedge significantly outperformed the VIX hedge in the first hedging
example, which is most directly related to small cap–specific risk. In the second example,
where perceived risk is macro in nature, the VIX hedge did better than the RVX hedge, but the
RVX hedge still enabled the hypothetical RVX-hedged Russell 2000 portfolio to deliver net
positive returns for the time period studied.
In conclusion, this case study suggests that when the risk to be hedged is specific to small
cap stocks, RVX futures–based hedges may provide for results superior to those of VIXbased hedges.
In the appendix below, the method of determining hedged performance is reviewed.
7
See, for example, the CNN Money reporting on that date: money.cnn.com/2014/07/31/investing/stocks-markets.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
7
Methodology appendix
This appendix explains how the examples in the simulation case study were constructed and
provides some relevant background information. The methodology used to construct, value
and liquidate hedges was designed to enable realistic comparison of RVX-based and VIXbased hedge strategies.
A1. Data and futures prices used in this study
This study covers the period January 1 to August 31, 2014. RVX and VIX index values,
contract prices, trade data and RVX quotes are obtained from CBOE, the Chicago Board
Options Exchange.8 Russell 2000 index total return values come from Russell Indexes.
The daily value of both RVX and VIX positions is determined by daily settlement prices. VIX
entry and exit trades are also based on settlement prices. Prices for VIX trades may be
obtained through the “trade at settlement” (TAS) transaction system available on the CBOE
Futures Exchange (CFE®), the trading venue for CBOE volatility futures. The prices assumed
for RVX trades and contract rolls are more conservatively estimated, because of the relative
thinness of trading. RVX positions are established based on the RVX settlement price but with
payment of a premium of one-half the average bid/ask spread quoted that day. This is an
estimated ask price. RVX positions are liquidated based on payment of the daily settlement
price minus one-half the average bid/ask price quoted that day, representing an estimated bid
price.
Contract rolls are assumed to take place at the average price ratio between the front-month
and second-month contract on the day of the roll. The roll ratio is defined as the price paid for
a second-month contract divided by the price at which the first-month contract is sold.
Average prices of VIX trades and average RVX bid and ask quotes are determined for each
15-minute block over the study period. VIX roll ratios are based on time-block-matched
average front-month and second-month contract trade prices. RVX roll ratios are based on
time-block-matched average ask prices for the second-month contract and bid prices for the
front-month contract. The average roll ratio on the day of the roll is the average of the ratios
defined for each 15-minute block.
A2. Sizing the hedges
A key implementation question the manager faces is how to size the hedge, regardless of
whether RVX or VIX contracts are used. For simplicity, we assume that a 50% hedge is
implemented. This means that the manager will implement a long volatility contract position
that starts off with a value that is 50% of the portfolio assets under management. Given that
RVX contract prices are typically higher than VIX contract prices, fewer RVX contracts will be
bought to implement similarly sized hedges.
The trader view of the size of a volatility futures position is usually in terms of the vega of the
position, which is the dollar change in position value with a 1-point change in the price of a
volatility contract.9 Vega can also be considered relative to the portfolio to be hedged. In this
case, the vega becomes equal to the hedge ratio (the hedge notional divided by value of the
underlying) divided by the contract price. With a hedge ratio of 50% and RVX prices at 20 and
VIX at 15, the respective vegas would be 3.33% and 2.50%. This means that a 1-point
increase in the VIX would add 3.33% to hedged portfolio value but a 1-point increase in RVX
would add only 2.5% to hedged portfolio value. Thus, equal notional sizes of the RVX and VIX
sizes will result in VIX providing a somewhat stronger hedge because of its higher vega. Once
8
VIX and RVX contract trading data can be found at cfe.cboe.com/Data/HistoricalData.aspx. VIX settlement data can also be obtained from
websites such as Quandl.com.
9
The vega of a single contract is equal to the contract multiplier. Both RVX and VIX contracts are priced at 1,000 times the transaction price, i.e.,
when a RVX contract is priced at 25, the price of a single contract is $25,000.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
8
a volatility contract position is implemented, the effective hedge ratio and vega will vary with
the value of the equity and futures positions.
A3. Contract prices and volatility index values
That hedge strategies must take account of the imperfect relationship between volatility
futures prices and spot volatility index values is an important aspect of this study. Volatility
futures contract prices are based on the expected value of the associated volatility index at
contract expiration. The more time there is until contract expiration, the more the price may
change along with the actual change in value of the volatility index.
Front-month futures contract prices have the closest relationship with volatility index values.
The front-month contract is the contract nearest to expiration. Figure 5 shows front-month
contract prices and underlying index values for RVX and VIX contracts over the study period.
It can be seen that contract prices are generally close to index values, but that there are
systematic differences.
Most of the time, the contract prices in Figure 5 are higher than the index values. This is often
referred to as “contango,” a trading term where the more specific meaning is that the contract
price is expected to decrease before expiration. On the other hand, it can also be seen that
the index value occasionally spikes considerably above the contract price. When the contract
price is below the index value, the contract can be said to be in “backwardation.” This term
has the more formal meaning that contract prices are expected to rise before expiration. In
volatility futures markets, however, backwardation is not common, and index values tend to
decrease more quickly than contract prices, pushing them back into contango before contract
expiration.
Figure 5 / Contract settlement prices and underlying index values for RVX and VIX
near contracts
January 1 to August 31, 2014
30
25
20
15
10
5
VIX
VIX front month contract settle
RVX
RVX front month contract settle
8/28/2014
8/14/2014
7/31/2014
7/17/2014
7/3/2014
6/19/2014
6/5/2014
5/22/2014
5/8/2014
4/24/2014
4/10/2014
3/27/2014
3/13/2014
2/27/2014
2/13/2014
1/30/2014
1/16/2014
1/2/2014
0
Source: CBOE.
A4. The RVX contract volatility premium
Since it is volatility futures contracts that can be traded, and not the index values themselves,
it is important to compare the RVX volatility premium as shown in Figures 2 and 4 with an
investable version of this premium based on front-month contract futures prices.
A comparison between the index-based premium and that based on front-month contract
futures prices is shown in Figure 6. The contract premium rises more slowly and is ultimately
lower than the rise in the index premium. It can also be seen that there are sometimes
Russell Indexes // RVX futures as a hedge for small cap–specific risk
9
significant variations in the short-term dynamics of the index and futures premia. In general,
the index premium can spike more sharply and quickly, while both premia can fall rapidly.
Figure 6 / Index and front-month contract price ratios for RVX and VIX
January 1 to August 31, 2014
1.8
1.7
1.6
1.5
1.4
1.3
1.2
1.1
RVX/VIX index ratio
8/28/2014
8/14/2014
7/31/2014
7/17/2014
7/3/2014
6/19/2014
6/5/2014
5/22/2014
5/8/2014
4/24/2014
4/10/2014
3/27/2014
3/13/2014
2/27/2014
2/13/2014
1/30/2014
1/16/2014
1/2/2014
1.0
RVX/VIX front month contract price ratio
Source: CBOE.
A5. Settlement prices and average daily contract prices and RVX trade prices
VIX average daily trading prices are typically close to settlement prices, which represent the
official end-of-day valuations of futures contracts. For VIX, the average difference between
average 15-minute block prices and settlement prices was $0.03 over the study period, with
a standard deviation of 0.39. On average, there is little difference between trading and
settlement prices.
RVX trading was not continuous enough over the study period for meaningful comparison of
settlement prices with daily trade prices. Instead, settlement prices may be compared to the
average of matched bid and ask quotes. Such implied prices are averaged in 15-minute
blocks and then averaged for each trading day. The implied price is the average of the bid
and ask. RVX contract implied prices track settlement prices about as well as the prices of
VIX contracts, with an average pricing difference of $0.04 and a standard deviation of 0.40.
The average RVX bid/ask spread over the study period is $0.32; its standard deviation is 0.07.
The daily average of this spread is used in the construction of RVX exit and entry prices.
A6. Construction of RVX and VIX hedge positions and estimation of their
performance
A long-lived hedge using futures contracts must be rolled on a periodic basis, due to contract
expirations. The CBOE volatility contracts used in this study are issued on, and expire on, a
monthly schedule. This roll strategy is based on the two nearest-to-expiration contracts and,
as already noted, the contracts are rolled on the last day of trading for the front-month
contract. This is the day before contract expiration. CBOE volatility futures contracts settle the
morning of contract expiration day.
Roll ratios are determined for all contract roll dates as described in the first section of this
appendix. If the contracts have the same price, then the roll ratio is 1. If the next month
contract is more expensive than the current month, as is common for volatility futures, the new
position will have fewer contracts, and the deflator will have to decrease in order to reflect that
in calculating the position value, the contract price is now multiplied by a smaller number of
contracts.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
10
These monthly roll ratios are used to create a roll deflator index series. This index starts out at
1.00. The value carries forward in time, except that the current value is divided by the roll ratio
on every roll date. At the end of August 2014, the values are 0.62 for the RVX roll deflator and
0.70 for VIX roll deflator. These values show that volatility contracts typically trade in
contango, and they demonstrate the substantial baseline cost of volatility-based hedging
strategies.
Indexes for hedge performance are then computed by multiplying front-month contract prices
by the hedge deflator on a day-by day basis. This multiplication adjusts the first-month
contract price for the effects of contract rolls on position value. The resulting indexes thus can
be used to determine the cumulative performance of the RVX-based and VIX-based hedge
strategies described here, once they are implemented. Figure 7 shows the performance of
these hedges over the study period.
Clearly, an investor would have lost money hedging continuously over the entire time period.
Over the entire study period, the average daily return to the RVX hedge strategy is -0.20%
and the standard deviation is 4.12%, compared to -0.13% and 4.22%, respectively, for the VIX
hedge. The relatively poorer overall performance of the RVX continuous hedge is in part the
result of basing RVX contract rolls on the ask price for the next month contract and the bid
price for the near contract. The superior performance of the RVX hedges in this case study
must thus be due to RVX relative performance during the periods hedged, and not to an
overall performance advantage.
Figure 7 / Performance of RVX and VIX continuous hedging indexes
130
120
110
100
90
80
70
60
RVX hedge continuous value
8/28/2014
8/14/2014
7/31/2014
7/17/2014
7/3/2014
6/19/2014
6/5/2014
5/22/2014
5/8/2014
4/24/2014
4/10/2014
3/27/2014
3/13/2014
2/27/2014
2/13/2014
1/30/2014
1/16/2014
50
1/2/2014
Hedge value on January 1, 2014 = 100
January 1, 2014 to August 31, 2014
VIX hedge continuous value
Source: Based on data from CBOE.
A7. Roll cost
Table 2 reports average roll spreads and standard deviations for RVX and VIX futures
contracts over the study period. The roll spread is the price difference between the frontmonth and second-month contracts at the time of contract roll, which is the day before
expiration of the front-month contract. The first set of rows reports averages and standard
deviations based on settlement prices. The second set of rows reports roll average spreads
and standard deviations based on average prices for VIX-based and average quote-based
prices for RVX. The quote-based price is based on the average of matched bids and asks for
both the front-month and second-month contracts. The third set of rows reports RVX roll
spreads based on the quote-based premia and discounts used in this study. Using settlement
prices, the average RVX roll cost is about 2.4% of hedge value. Utilizing the quote spread
increases the average RVX roll costs to about 5.1% of hedge value.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
11
Table 2 / Average roll spreads for RVX and VIX volatility contracts
January 1, 2014 to August 31, 2014
RVX
VIX
Settlement price–based
0.45
0.69
Standard deviation
0.55
0.41
Average price or quote based
0.63
0.67
Standard deviation
0.55
0.35
Average bid/ask roll spread
0.96
---
Standard deviation
0.57
---
Data source: CBOE trade data.
A8. Performance Accounting
Some specific details regarding the construction and rolling of hedges may be helpful in
providing for a more intuitive understanding of the results. Entry and exit and daily valuation
are treated first, followed by a roll example.
Consider that the value of a passive Russell 2000 portfolio is 100 on January 1, 2014, and
101.32 at market close on February 21, 2014. On this latter date, a futures position is
established for the first hedging example with a target value of 50% of portfolio value, or
50.66. The average quote spread is 1.76% for RVX front-month contracts, so the cost of
establishing the RVX position is assumed to be 0.88% of the contract value. The returns of
the hedge portfolios are based on the continuous value indexes described in Section A6 of
this appendix. On February 21, 2014, the settlement price of the RVX front-month contract is
19.65. On February 24, the settlement price is 19.45. The value of the position thus declines
by 1.03%. The total return on the hedge portfolio at day-end February 24 is thus
-1.90% = (1 – 0.088)*(1-.013) - 1.
The Russell 2000 closes at 102.18 on February 24 for a daily total return of 0.85%. The total
return of the hedged portfolio is then
-0.01% = 0.85% + ½ * (-1.90%);
that is, the equity portfolio return plus the return on the hedge portfolio times its starting value
relative to the value of the equity portfolio. On subsequent days, the return of the hedge
portfolio must be multiplied by the closing value on the previous day of the hedge portfolio as
a fraction of the equity portfolio. The cost of exiting an RVX hedge is calculated analogously to
the way entry is shown here.
For the VIX contract, on February 21, 2014, the settlement price of RVX is 15.15. On
February 24, the settlement value is 15.00, a 1.00% decline in value. As there is no cost to
establish the VIX position, the return of the VIX-hedged portfolio for the day is
0.36% = 0.85% + ½ * (-1.00%).
Contract rolls involve the exchange of front-month contracts about to expire for next-month
contracts. The old futures position is liquidated, and the proceeds are used to establish the
new futures position. If the roll ratio is 1.05 for a month and the hedge consists of 100 frontmonth contracts, the number of contracts after the roll will be
95.24 = 100 / 1.05,
so that the value of the new position will be equal to the value of the old position at the time of
trade. It is assumed that the real position sizes are large enough such that fractional contracts
do not trade. The daily value of the new position is based on the daily settlement price. The
roll deflator properly adjusts portfolio value for the change in the number of contracts in the
hedge portfolio.
Russell Indexes // RVX futures as a hedge for small cap–specific risk
12
About Russell Indexes
Russell’s indexes business, which began in 1984, accurately measures U.S. market segments
and tracks investment manager behavior for Russell’s investment management and
consulting businesses. Today, our series of U.S. and global equity indexes reflect distinct
investment universes – asset class, geographic region, capitalization and style – with no gaps
or overlaps. Russell Indexes offers more than three dozen product families and calculates
more than 700,000 benchmarks daily, covering 98% of the investable market globally, 81
countries and more than 10,000 securities. As of December 31, 2014, approximately $5.7
trillion in assets are benchmarked to the Russell Indexes.
For more information about Russell Indexes, call us or visit www.russell.com/indexes.
Americas: +1-877-503-6437; APAC: +65-6880-5003; EMEA: +44-0-20-7024-6600
Disclosures
© 2015 London Stock Exchange Group companies.
London Stock Exchange Group companies includes FTSE International Limited (“FTSE”), Frank Russell Company (“Russell”), MTS
Next Limited (“MTS”), and FTSE TMX Global Debt Capital Markets Inc (“FTSE TMX”). All rights reserved.
“FTSE®”, “Russell®”, “MTS®”, “FTSE TMX®” and “FTSE Russell” and other service marks and trademarks related to the FTSE or
Russell indexes are trade marks of the London Stock Exchange Group companies and are used by FTSE, MTS, FTSE TMX and
Russell under licence.
All information is provided for information purposes only. Every effort is made to ensure that all information given in this publication
is accurate, but no responsibility or liability can be accepted by the London Stock Exchange Group companies nor its licensors for
any errors or for any loss from use of this publication.
Neither the London Stock Exchange Group companies nor any of their licensors make any claim, prediction, warranty or
representation whatsoever, expressly or impliedly, either as to the results to be obtained from the use of the FTSE Russell Indexes
or the fitness or suitability of the FTSE Russell Indexes for any particular purpose to which they might be put.
The London Stock Exchange Group companies do not provide investment advice and nothing in this document should be taken as
constituting financial or investment advice. The London Stock Exchange Group companies make no representation regarding the
advisability of investing in any asset. A decision to invest in any such asset should not be made in reliance on any information
herein. Indexes cannot be invested in directly. Inclusion of an asset in an index is not a recommendation to buy, sell or hold that
asset. The general information contained in this publication should not be acted upon without obtaining specific legal, tax, and
investment advice from a licensed professional.
No part of this information may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise, without prior written permission of the London Stock Exchange Group
companies. Distribution of the London Stock Exchange Group companies’ index values and the use of their indexes to create
financial products require a licence with FTSE, FTSE TMX, MTS and/or Russell and/or its licensors.
Past performance is no guarantee of future results. Charts and graphs are provided for illustrative purposes only. Index returns
shown may not represent the results of the actual trading of investable assets. Certain returns shown may reflect back-tested
performance. All performance presented prior to the index inception date is back-tested performance. Back-tested performance is
not actual performance, but is hypothetical. The back-test calculations are based on the same methodology that was in effect when
the index was officially launched. However, back-tested data may reflect the application of the index methodology with the benefit of
hindsight, and the historic calculations of an index may change from month to month based on revisions to the underlying economic
data used in the calculation of the index.
First use: June 2015
CORP-10465-06-2017
Russell Indexes // RVX futures as a hedge for small cap–specific risk
13