The VIX Compass: August 28xa.yimg.com/.../113602213/name/Barclays_The_VIX_Compass_Augu… ·...

Equity Research28 August 2013

Barclays Capital Inc. and/or one of its affiliates does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report.

Investors should consider this report as only a single factor in making their investment decision.

This research report has been prepared in whole or in part by equity research analysts based outside the US who are not registered/qualified as research analysts with FINRA.

PLEASE SEE ANALYST CERTIFICATION(S) AND IMPORTANT DISCLOSURES BEGINNING ON PAGE 38.

The VIX Compass

August 28 In this weekly publication, we give a comprehensive overview of the VIX and VSTOXX derivatives landscape. The aim is to provide value-added market intelligence covering VIX and VSTOXX ETPs, futures, and options. This publication serves not only as a forum to monitor the performance of several hedging and alpha strategies that we have previously analyzed, but also as a cross-regional perspective to identify relative value between the regions.

Market Commentary • Over the past week, the S&P 500 declined 1.3% and the VIX futures increased

across the curve with the front end rising more than the back end. The September, November and December VIX futures increased by 1.55, 0.95 and 0.75 points, respectively.

• The VIX two-month implied volatility increased 5 points to 87% and VIX two-month realized volatility increased at 52% over the past week. The VIX implied volatility trades above our model’s fair value by 17 points. The VIX 2M skew stayed declined 8 points to 36% and is trading rich compared to our model skew by 4 points. The VIX implied volatility term structure 3M/1M volatility ratio declined sharply to 0.766, with the one-year average at 0.86 last week.

• The V2X 2M implied volatility stayed flat around 65% and is trading cheaper than the theoretical model value by 12 points over the past week. The V2X 2M realized also increased by 12 points to 60%.

• The Short-term VIX Futures Index (SPVXSTR) roll cost decreased sharply to 3.66%. The Medium-term VIX Futures Index (SPVXMTR) roll cost stayed declined to 2.6%.

• Over the past week, the SX5E declined 1.4% and the VSTOXX increased 3 points. The implied roll costs for VST1MT declined to -1% and VMT5MT decreased to 5%.

DERIVATIVES

U.S. Equity Derivatives Strategy

Global & U.S. Equity Derivatives Strategy Maneesh S. Deshpande 1.212.526.2953 [email protected] BCI, New York

U.S. Equity Derivatives Strategy Rohit Bhatia 1.212.526.0367 [email protected] BCI, New York

Ashish Goyal 1.212.526.2771 [email protected] BCI, New York

Arnab Sen 1.212.526.5429 [email protected] BCI, New York

European Equity Derivatives Strategy Christian Kober +44 (0)20 3134 8673 [email protected] Barclays, London Anshul Gupta +44 (0)20 3134 8122 [email protected] Barclays, London Kim Berg +44 (0)20 3555 0289 [email protected] Barclays, London

Link to US VolCenter

Vol Center is a premier Equity Derivatives application providing Market Data, Analytics, Market Insight and Research

Sources for all charts and tables: Barclays Research, OptionMetrics, Bloomberg, unless otherwise noted For strategies in the report: Past performance is not a guarantee of future results

https://live.barcap.com/go/BC/barcaplive?menuCode=MENU_EQ_DERIV_USV�

Barclays | The VIX Compass

28 August 2013 2

CONTENTS

PERFORMANCE OF ALPHA AND HEDGING STRATEGIES ...................... 3

VIX LIQUIDITY LANDSCAPE ........................................................................... 4

VIX ETP LIQUIDITY ............................................................................................ 5

VIX FUTURES AND OPTIONS LIQUIDITY ..................................................... 6

VIX ETP SNAPSHOT .......................................................................................... 7

VSTOXX LIQUIDITY LANDSCAPE .................................................................. 8

VIX ETP AND FUTURES .................................................................................... 9

VSTOXX ETPS AND FUTURES ...................................................................... 10

VIX AND VIX ETP OPTIONS .......................................................................... 11

VIX OPTION EXPIRATION SNAPSHOT ....................................................... 12

VSTOXX OPTIONS .......................................................................................... 13

VIX AND VSTOXX – RELATIVE VALUE ....................................................... 14

US HEDGING STRATEGIES ............................................................................ 15

EU HEDGING STRATEGIES ............................................................................ 17

ALPHA STRATEGIES ....................................................................................... 18

MARKET INTELLIGENCE ................................................................................ 21

US HEDGING STRATEGIES: METHODOLOGY .......................................... 29

EU HEDGING STRATEGIES: METHODOLOGY .......................................... 33

ALPHA STRATEGIES: METHODOLOGY ..................................................... 35


28 August 2013 3

PERFORMANCE OF ALPHA AND HEDGING STRATEGIES

Performance of Systematic Alpha Strategies • The short SPVXSTR/long SPVXMTR strategy decreased at 1% last week, as short-term

VIX futures increased more than the mid-term VIX futures.

• The long VST1MT short SPVXSTR strategy declined 2.5% as the VIX futures front end increased more than the V2X futures front end.

• The VIX 1M 80%-100% 1x2 put spread dropped 10.6% as the August VIX futures settled at 16.42 from 14.91(8/20) at the expiration and the strategy was short delta.

• The VIX delta hedged 1M ATM straddles returned a small loss as the September VIX future realized volatility picked up in the past week.

Performance of Systematic Hedging Strategies • SPX Hedges: The hedges had again had mixed performance as compared to an S&P 500

portfolio which dropped 1.3%. The best hedge was the SPVXSP due to the sharp rise in front month volatilities in the past week.

• SX5E Hedges: Over the past week, the SX5E Put Spread 12M Collar was the best hedge, outperforming other strategies as the SX5E declined 1.4%.


28 August 2013 4

VIX LIQUIDITY LANDSCAPE

FIGURE 1 Total Vega Outstanding ($ Millions)

0

500

1000

1500

2000

2500

3000

0

200

400

600

800

1,000

1,200

Aug-12 Sep-12 Oct-12 Nov-12 Dec-12 Jan-13 Feb-13 Mar-13 Apr-13 May-13 Jun-13 Jul-13 Aug-13

VIX Futures VIX Options VIX ETP Listed SPX & SPY Options (RHS)

Total Vega Outstanding is the weekly moving average of the daily vega outstanding. VIX ETP Vega Outstanding is calculated by finding Gross Vega outstanding in the ETPs covered in Figure 14 for which we have the calculated Vega ($MM). Prices as of 08/27/2013.

FIGURE 2 Total Vega Volume ($ Millions)

0

100

200

300

400

500

600


VIX Futures VIX Options VIX ETP Listed SPX & SPY Options

Total Vega Volume is the weekly moving average of the daily vega volumes traded. VIX ETP Vega Volume is calculated by finding Gross Vega Volume in the ETPs covered in Figure 14 for which we have the calculated Vega ($MM). Prices as of 08/27/2013.


28 August 2013 5

VIX ETP LIQUIDITY

FIGURE 3 Current Gross VIX ETP AUM by Type

FIGURE 4 Current gross VIX ETP Vega AUM by Tenor

Unlevered 46%

Leveraged36%

Dynamic15%

0

20

40

60

80

100

120

140

160

180

Short-Term Mid Term

Short Vega Exposure Long Vega Exposure Short Interest Vega

Vega($MM)

Details on the different ETP Types in the Appendix Details on the ETP Tenors in the Appendix

FIGURE 5 Gross VIX ETP Vega AUM by Type

FIGURE 6 Rolling 1 Month VIX ETP Vega flow by Type

0

50

100

150

200

250

300

350

Aug-12 Nov-12 Feb-13 May-13 Aug-13

Unleveraged Leveraged Dynamic

Vega($MM)

-60

-40

-20

0

20

40

60

80

100

120


Unleveraged Leveraged Dynamic

Vega($MM)

Details on the different ETP Types in the Appendix Details of ETP Flow Calculation is shown in the Appendix

FIGURE 7 VIX ETP Vega by direction of volatility exposure

0

50

100

150

200

250

300

350

400

Aug-12 Oct-12 Dec-12 Feb-13 Apr-13 Jun-13

Long Exposure Short Exposure Short Interest

Vega($MM)

FIGURE 8 Theoretical Flow from Leveraged ETP rebalancing

-80-60-40-20

020406080

100120

-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6

Thou

sand

s

Leveraged ETP Flows 22D MA VIX Future Volume

VIX Future Contracts

Move in 1M VIX Future

Details of the theoretical flow calculation for leveraged ETPs in the Appendix


28 August 2013 6

VIX FUTURES AND OPTIONS LIQUIDITY

FIGURE 9 VIX futures and VIX ETP short-term Vega AUM

FIGURE 10 VIX futures and VIX ETP mid-term Vega AUM

0%

20%

40%

60%

80%

100%

120%

0

50

100

150

200

250

300

350


Gross ETP Short-termVIX Futures Short-termShort-term ETP/Short-term VIX Futures (RHS)

Vega($MM) Ratio

0%

5%

10%

15%

20%

25%

30%

35%

40%

0

50

100

150

200

250

Aug-12 Nov-12 Feb-13 May-13 Aug-13Gross ETP Mid-termVIX Futures Mid-termMid-term ETP/Mid-Term VIX Futures (RHS)

Vega($MM)

Gross short-term vega aggregates all ETP vega in the first 2 month VIX Futures Gross mid- term vega aggregates all ETP vega in the 3-7 month VIX Futures

FIGURE 11 VIX Futures vega AUM versus SPX & SPY vega

0%

5%

10%

15%

20%

25%

30%

0

500

1,000

1,500

2,000

2,500

3,000


VIX Futures SPX + SPY Listed VIX Futures/SPX (RHS)

RatioVega($MM)

FIGURE 12 VIX and SPVXSTR Options Open Interest

0%

10%

20%

30%

40%

50%

60%

0

50

100

150

200

250

300

Aug-12 Oct-12 Dec-12 Feb-13 Apr-13 Jun-13VIX Option SPVXSTR Options

VIX Put/Call OI (RHS)

Ratio

FIGURE 13 VIX and SPVXSTR Options Volume

0%

20%

40%

60%

80%

100%

0

10

20

30

40

50


VIX Option SPVXSTR Options

VIX Put/Call Volume (RHS)

Ratio


28 August 2013 7

VIX ETP SNAPSHOT

FIGURE 14 Comparison of various VIX ETPs

Ticker Type Maturity AUM VolumeShort

Interest*

Rolling 1 Month Flow

Vega Per

Share

($mm)Long

ExposureShort

ExposureVega($m

m)Vega($m

m)Vega($m

m)Futures

Sells2Futures

Buys2

VIX Futures

- 383 - 248 - - - - -

VIX Options

- - 214 - 22 - - - - -

All ETPs All ETPs All ETPs 3730 180 37 171 90 -4 - -7.0 6.8

VXX UnLeveraged Short 1504 87 0 73 71 1 1.0 -4.4 4.2

VQT Dynamic Short 574 1 0 0 - 0 0.2 - -

UVXY Leveraged Short 318 34 0 44 7 -4 4.8 -1.9 1.8

XIV Leveraged Short 268 0 18 33 0 1 1.6 0.8 -0.7

VIXY UnLeveraged Short 194 11 0 3 1 1 2.6 -0.6 0.5

MHUU Dynamic Mixed 176 8 9 0 - -5 1279.5 - -

XVZ Dynamic Mixed 171 7 2 0 - 0 2.2 - -

TVIX Leveraged Short 167 18 0 5 4 1 0.2 -1.0 0.9

VIXM UnLeveraged Mid 98 5 0 0 0 2 1.3 -0.1 0.1

SVXY Leveraged Short 71 0 5 9 0 -1 6.4 0.2 -0.2

VXZ UnLeveraged Mid 59 3 0 1 5 0 1.0 -0.1 0.0

ZIV Leveraged Mid 48 0 3 0 0 1 1.8 0.0 0.0

VIIX UnLeveraged Short 17 1 0 1 0 0 0.5 0.0 0.0

HVU Leveraged Mixed 15 2 0 1 2 0 0.5 -0.1 0.1

XVIX Dynamic Mixed 14 1 0 0 - 0 1.6 - -

Others Others Others 37 2 0 0 0 0 31.3 0.0 0.0

Vega1($mm)Theoretical Flow due to daily roll

Vega($mm)

Source: Barclays Research, Bloomberg. 1Vega is Vega outstanding for the ETP. For Dynamic ETPs the Vega is computed by adding the vega of the sub component indices. 2 Futures Sells and Buys are the expected gross aggregate futures vega($mm) to be sold and bought in rebalancing the ETP. *Short Interest is shown till the latest reported date.


28 August 2013 8

VSTOXX LIQUIDITY LANDSCAPE

FIGURE 15 Vega outstanding in VSTOXX products

0

500

1000

1500

2000

2500

3000

0

5

10

15

20

25

30

35

40

45

Dec-10 Mar-11 Jun-11 Sep-11 Dec-11 Mar-12 Jun-12 Sep-12 Dec-12 Mar-13 Jun-13

Mill

ions

Vega Outstanding

Mill

ions

V2X Options V2X Futures Listed SX5E (RHS)

FIGURE 16 Vega volume in VSTOXX products

0

10

20

30

40

50

60

70

80

90

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

Dec-10 Mar-11 Jun-11 Sep-11 Dec-11 Mar-12 Jun-12 Sep-12 Dec-12 Mar-13 Jun-13

Mill

ions

Vega Volume

Mill

ions

V2X Options V2X Futures Listed SX5E (RHS)

FIGURE 17 VSTOXX Options Open Interest

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

0

50,000

100,000

150,000

200,000

250,000


Contracts

V2X OI Ratio(RHS) V2X Call V2X Put

FIGURE 18 VSTOXX Options Volume

0

5

10

15

20

25

30

0

5,000

10,000

15,000

20,000

25,000

30,000


Contracts

V2X Volume Ratio(RHS) V2X Call V2X Put

V2X OI Ratio = (Put OI – Call OI)/(Put OI + Call OI) V2X OI Ratio = (Put Volume – Call Volume)/(Put Volume + Call Volume)


28 August 2013 9

VIX ETP AND FUTURES

FIGURE 19 Performance of VIX Futures, SPVXSTR and SPVXMTR

FIGURE 20 Estimated SPVXSTR & SPVXMTR Monthly Roll Cost

0

20

40

60

80

100

120


F1 F5 SPVXSTR SPVXMTR

-10%

-5%

0%

5%

10%

15%

20%


SPVXSTR SPVXMTR

F1 & F5 are constant maturity 1 & 5 month VIX futures. Roll cost is estimated using the term structure premium in VIX futures.

FIGURE 21 VIX Futures Term & Open Interest Snapshot

FIGURE 22 Hedge Ratio of VIX Futures versus SPX

10

12

14

16

18

20

22

020406080

100120140160180200

Sep-13 Nov-13 Jan-14 Mar-14 May-14

Cont

ract

s('0

00)

OI Today(LHS) Future Today Future Wk Ago

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

1M 2M 3M 4M 5M 6M

SPV

XST

R

SPV

XM

TR

Hedge Ratios are calculated using past 2 year overlapping weekly returns.

FIGURE 23 Scatter plot of 1M SPVXSP versus 1M SPX returns

FIGURE 24 Scatter plot of 1M SPVXMP versus 1M SPX returns

-60%

-40%

-20%

0%

20%

40%

60%

-10% -5% 0% 5% 10% 15%

SPVXSP(y axis) vs SPX (x axis)SPVXSP vs SPX Last 2 weeks

-30%

-20%

-10%

0%

10%

20%

30%

-10% -5% 0% 5% 10% 15%

SPVXMP(y axis) vs SPX (x axis)

SPVXMP vs SPX Last 2 weeks Plots are calculated using past 2 year overlapping monthly returns. Plots are calculated using past 2 year overlapping monthly returns.


28 August 2013 10

VSTOXX ETPS AND FUTURES

FIGURE 25 Performance of VSTOXX Futures, VST1MT and VMT5MT

FIGURE 26 Estimated VST1MT & VMT5MT Monthly Roll Cost

152535455565758595

105115


VSTOXX Future 1M VSTOXX Future 5M

VST1MT VMT5MT

-10%

-5%

0%

5%

10%

15%


VST1MT VMT5MT

Prices are normalized to 100 at the starting date to compare the performance. Roll cost is estimated using the term structure premium in VSTOXX futures.

FIGURE 27 VSTOXX Futures Term & Open Interest Snapshot

FIGURE 28 Hedge Ratio of VSTOXX Futures versus SX5E

0

10

20

30

40

50

60

17

18

19

20

21

22

23

24

Spot 1M 2M 3M 4M 5M

Con

trac

ts(0

00)

OI Today (RHS) Future Today Future 1 Wk Ago

0%

10%

20%

30%

40%

50%

60%

70%

80%

1M 2M 3M 4M 5M 6M

VST

1MT

VM

T5M

T

Hedge Ratios are calculated using past 2 year overlapping weekly returns.

FIGURE 29 Scatter plot of 1M VST1MT versus SX5E returns

FIGURE 30 Scatter plot of 1M VMT5MT versus SX5E returns

-40%

-20%

0%

20%

40%

60%

-15% -10% -5% 0% 5% 10% 15% 20% 25%

VST1MT(y axis) vs SX5E( x axis)VST1MT vs SX5E Last 2 weeks

-30%

-20%

-10%

0%

10%

20%

30%

-20% -10% 0% 10% 20% 30%

VMT5MT(y axis) vs SX5E (x axis)VMT5MT vs SX5E last 2 weeks

Plots are calculated using past 2 year overlapping monthly returns. Plots are calculated using past 2 year overlapping monthly returns.


28 August 2013 11

VIX AND VIX ETP OPTIONS

FIGURE 31 2M VIX, 2M VIX Theoretical and 1M SPVXSTR ATM Volatility

FIGURE 32 VIX 2M IV/RV and SPVXSTR 1M IV/RV ratio

40%

50%

60%

70%

80%

90%

100%

110%


VIX 2M 50D IV SPVXSTR 1M 50D IV

VIX 50Delta Theo IV

40%

90%

140%

190%

240%

290%

340%


VIX 2M IV/RV SPVXSTR 1M IV/RV

Theoretical VIX Implied Volatility Calculation can be found in the Appendix Details of VIX RV calculation in the Appendix

FIGURE 33 2M VIX, 2M VIX Theoretical and 1M SPVXSTR Skew

FIGURE 34 VIX 2M and SPVXSTR 1M Skew Snapshot

0%

10%

20%

30%

40%

50%

60%

70%

80%


VIX 2M Skew SPVXSTR 1M Skew

VIX 2M Theo Skew

40%

50%

60%

70%

80%

90%

100%

110%

25 30 35 40 45 50 55 60 65 70 75

VIX 2M IV SPVXSTR 1M IV

Skew= (25D Call Vol)/(50D Call Vol) - (25D Put Vol)/(50D Put Vol) Volatility for standardized put deltas

FIGURE 35

VIX ATM Volatility and Open Interest Term Structure Snapshot

FIGURE 36 VIX 3M/1M ATM Volatility Ratio

50%

60%

70%

80%

90%

100%

110%

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

Sep-13 Oct-13 Nov-13 Dec-13 Jan-14 Feb-14

Con

trac

ts(0

00)

Expiry

Open Interest Today (LHS)VIX ATM IV TodayVIX ATM IV 1 Wk Ago

40%

50%

60%

70%

80%

90%

100%

110%


3M/1M VIX Vol

VIX 50 delta implied volatility for different maturities VIX 50 delta implied volatility for 3M and 1M


28 August 2013 12

VIX OPTION EXPIRATION SNAPSHOT

FIGURE 37 Expiring Open Interest Notional Distribution By Strike

FIGURE 38 Dollar Gamma Distribution By Strike

0

100

200

300

400

500

600

700

800

9 11 13 15 17 19 21 23 25 27 29 32 37 42 47 55 65 75

VIX Future Level

Open Interest Notional ($MM)

02468

101214161820

9 11 13 15 17 19 21 23 25 27 29

VIX Future Level

Dollar Gamma ($MM)

Open Interest is only for upcoming Expiration Dollar Gamma = Change in Dollar Delta for 1% move in underlying

FIGURE 39 Historical Open Interest for most liquid strikes

FIGURE 40 Open Interest distribution by Expirations

10

15

20

25

30

35

40

050,000

100,000150,000200,000250,000300,000350,000400,000

Mar-13 May-13 Jul-13

14 17 16 price (RHS)

Contracts September13 VIX

55%

29%

9%5%

2% 0%0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

Sep-13

Oct-13

Nov-13

Dec-13

Jan-14

Feb-14

Open Interest Notional($MM)

First 3 lines in the legend denote the most active strikes


28 August 2013 13

VSTOXX OPTIONS

FIGURE 41 2M VSTOXX and 2M VSTOXX Theoretical ATM Volatility

FIGURE 42 VSTOXX and SX5E 2M IV/RV ratio

30%

40%

50%

60%

70%

80%

90%

Aug-12 Nov-12 Feb-13 May-13 Aug-13V2X 2M 50D Theo IV V2X 2M 50D IV

0%20%40%60%80%

100%120%140%160%180%


Ratio

V2X 2M IV/RV Ratio SX5E 2M IV/RV Ratio Theoretical VSTOXX Implied Volatility calculation can be found in the Appendix Details of VSTOXX RV calculation in the Appendix

FIGURE 43 VSTOXX Term Structure (3M/1M ATM Volatility Ratio)

FIGURE 44 2M VSTOXX Skew

20%

30%

40%

50%

60%

70%

80%

90%

100%

110%


3M/1M V2X Vol

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%


V2X 2M Skew V2X 50 delta implied volatility for 3M and 1M Skew= (25D Call Vol)/(50D Call Vol) - (25D Put Vol)/(50D Put Vol)

FIGURE 45

VSTOXX/SX5E 2M Implied Volatility Ratio

FIGURE 46 Expiring Open Interest Notional Distribution By Strike

1.4

1.9

2.4

2.9

3.4

3.9

4.4

4.9

Aug-12 Nov-12 Feb-13 May-13

V2X/SX5E 2M IV Ratio

0

5

10

15

20

25

30

35

17 19 21 23 25 27 29

V2X Future Level

Open Interest Notional ($MM)

Open Interest is only for upcoming Expiration


28 August 2013 14

VIX AND VSTOXX – RELATIVE VALUE

FIGURE 47 VIX and VSTOXX Futures Term Structure

FIGURE 48 VIX versus VSTOXX Constant Maturity 1M and 5M Futures

-25%

-20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

0

5

10

15

20

25

Spot 1M 2M 3M 4M 5M

%Diff between VIX Fut %Diff between V2X Fut

V2X level (LHS) VIX Level(LHS)

-10123456789

Aug-12 Oct-12 Dec-12 Feb-13 Apr-13 Jun-13 Aug-13

VSTOXX - VIX Const Mat 1M Futures Spread

VSTOXX - VIX Const Mat 5M Futures Spread

% difference refers to the difference between the adjacent contracts (Fi/Fi-1-1)

FIGURE 49 SPVXSTR and VST1MT Monthly Roll Cost

FIGURE 50 SPVXMTR and VMT5MT Monthly Roll Cost

-10%

-5%

0%

5%

10%

15%

20%


VST1MT SPVXSTR

-4%

-2%

0%

2%

4%

6%

8%


VMT5MT SPVXMTR

Roll cost is estimated using the term structure premium in VIX and VSTOXX Roll cost is estimated using the term structure premium in VIX and VSTOXX mid-

FIGURE 51 VIX and VSTOXX 2M Implied Volatility

FIGURE 52 VIX and VSTOXX 2M IV/RV Ratio

40%

50%

60%

70%

80%

90%

100%

110%

Apr-12 Jul-12 Oct-12 Jan-13

V2X 2M 50D IV VIX 2M 50D IV

40%

60%

80%

100%

120%

140%

160%

180%

Apr-12 Jul-12 Oct-12 Jan-13

V2X 2M IV/RV Ratio VIX 2M IV/RV


28 August 2013 15

US HEDGING STRATEGIES

FIGURE 53 SPVXSP & SPVXMP Excess Return(ER)

FIGURE 54 Dynamic VIX(SPDVIXP), BDVIX & LDVIX ER

85

90

95

100

105

110

115

Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13

SPVXSP SPVXMP

80

90

100

110

120

130

140


SPDVIXP BDVIX LDVIX SPVXSP and SPVXMP rebalanced on VIX expirations. BDVIX is Backwardation Dynamic VIX and LDVIX is Long only Dynamic VIX

FIGURE 55 VIX call and VIX Call Options +Short DH Straddle ER

FIGURE 56 VIX Risk Reversals ER

80

85

90

95

100

105

110

115

120

125

Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13VIX 3M 120% Call OptionVIX 1M Call Option + Short DH VIX Straddle

60

70

80

90

100

110

120


VIX Risk Reversal 4xSPVXMP- 1xSPVXSP

VIX 1M 120% Call + VIX 1M Short Delta Hedged Straddle Notional amount of SPVXSP is one-fourth of SPVXMP notional amount

FIGURE 57 Short VIX 1x2 Call Spreads and VIX 1M-3M Strangle ER

FIGURE 58 12M Rolling Put Spread Collar ER

80

85

90

95

100

105

110


VIX 1x2 Call Spread VIX 1M-3M 95% - 120% Strangle

92

94

96

98

100

102

104

106

108


1Y 40-15-25 Delta SPX Put Spread Collar

VIX 3M 1x2 120%-150% call spread & VIX 1M-3M 95%-120% Strangle. 12M Put spread with 15-40 delta strikes and 1M Call with 25 delta strikes.

Note : Past performance is not a guarantee of future results.


28 August 2013 16

FIGURE 59 Performance Comparison of Hedging Strategies since 2007

StrategiesHedge Notional

Average Return

Standard Deviation

Sharpe Ratio

Sortino Ratio

Draw-down

1 Week Return

3 Month Return

SPX 7.1% 24.0% 0.25 0.34 -55.3% -1.3% -0.6%

SPX + VIX 3M 120% CallOption

12.2% 6.6% 21.9% 0.25 0.35 -50.9% -1.1% -1.0%

SPX + VIX 1M 80%-120% Risk Reversal

5.2% 6.1% 21.8% 0.23 0.32 -50.7% -1.1% -1.2%

SPX + VIX 3M 120% - 150% 1x2Call Spread

19.7% 5.4% 22.8% 0.19 0.26 -53.5% -1.4% -1.2%

SPX + VIX 1M 120% Call + Short VIX 1M ATM DH Straddle

11.1% 7.5% 21.6% 0.30 0.41 -48.0% -1.2% -1.2%

SPX + VIX 1M-3M Strangle

7.1% 7.4% 24.0% 0.26 0.36 -54.8% -1.6% -1.5%

SPX + SPVXSP 3.3% 6.6% 22.1% 0.25 0.35 -51.6% -1.0% -0.6%

SPX + SPVXMP 7.1% 7.0% 22.0% 0.27 0.38 -50.9% -1.0% -0.6%

SPX + long SPVXMP & short SPVXSP

17.9% 7.6% 22.2% 0.29 0.41 -50.6% -1.1% -0.4%

SPX + Dynamic VIX (SPDVIXP)

16.5% 10.9% 20.7% 0.48 0.68 -42.0% -1.4% -0.4%

SPX + BDVIX 16.5% 10.5% 21.4% 0.44 0.63 -44.6% -1.3% -0.5%

SPX + LDVIX 16.5% 9.9% 20.3% 0.43 0.62 -42.9% -1.1% -0.5%

SPX + SPX rolling 12M Put Spread Collar

25.1% 6.8% 21.9% 0.26 0.36 -51.3% -1.0% -0.1%

Note: Hedges are sized so that the 75% monthly VaR cost of the hedges is 5%/year. Data as of 8/27/13. Past performance is not a guarantee of future results. Start date of the strategies is selected to keep in line with the VIX option inception date.


28 August 2013 17

EU HEDGING STRATEGIES

FIGURE 60 Excess Return V2X based Investable Vol Strategies

FIGURE 61 SX5E Hedged with V2X Based Strategies

0

50

100

150

200

250

300

350

400

450

500


VST1ME VMT5ME Dynamic VSTOXX

30

50

70

90

110

130


SX5E SX5E + VST1MT

SX5E + VMT5MT SX5E + Dynamic VSTOXX

Past performance is not a guarantee of future results. Past performance is not a guarantee of future results.

FIGURE 62 SX5E Options-based Excess Return Hedges

FIGURE 63 SX5E Hedged with Vanilla Options-Based Hedges

40

50

60

70

80

90

100

110

120

130

140


Put 1M Put Spread Collar 12M

30

40

50

60

70

80

90

100

110

120


SX5E SX5E + Put 1M SX5E + PS Collar 12MPast performance is not a guarantee of future results. Past performance is not a guarantee of future results.

FIGURE 64 Performance Comparison of Hedging Strategies Since 2007

StrategiesHedge

NotionalAverage Return

Standard Deviation

Sharpe Ratio

Sortino Ratio

Draw-down

1 Week Return

3 Month Return

SX5E 1.4% 26.6% 0.00 0.00 -58.2% -1.4% -0.9%

SX5E + VST1MT 3.5% 1.7% 24.8% 0.01 0.01 -53.9% -1.2% -0.8%

SX5E + VMT5MT 7.5% 2.0% 25.6% 0.02 0.03 -53.9% -1.3% -0.9%

SX5E + Dynamic VSTOXX 13.5% 5.1% 25.4% 0.14 0.20 -48.5% -1.4% -1.1%

SX5E + Put1M 31.0% 0.8% 23.7% -0.03 -0.04 -56.2% -1.2% 0.7%

SX5E + PutSpreadCollar12M 25.0% 2.0% 24.1% 0.02 0.03 -54.1% -1.2% -0.1%

Note: Hedges are sized so that the 75% monthly VaR cost of the hedges is 5%/year. Data as of 8/27/13. Past performance is not a guarantee of future results. Start date of the strategies is selected to keep in line with the VIX option inception date.


28 August 2013 18

ALPHA STRATEGIES

FIGURE 65 Short SPVXSTR & Long SPVXMTR and Leveraged Alpha Dynamic VIX

FIGURE 66 Short Delta Hedged 1M ATM VIX Straddles

0

100200

300

400

500

600

700

800900

1,000


SPVXSTR versus SPVXMTR Leveraged Alpha Dynamic VIX

0

50

100

150

200

250


Delta Hedged 1M ATM VIX Straddle

Long SPVXMTR amount is 100% of equity while short SPVXSTR is 50%. 0.15 vega per $100 of equity is sold through delta-hedged straddle.

FIGURE 67 1M 80%-100% 1x2 VIX Put Spread

FIGURE 68 Long 1M ATM VIX Put & Short 2M Same Strike Put

0

50

100

150

200

250


VIX 1M 80%-100% 1x2 Put Spread

020406080

100120140160180200


VIX Long 1M ATM Put and Short 2M Same Strike Put

Dollar notional of 1M ATM VIX put is 100% of equity. Dollar notional of both long and short puts is 100% of portfolio equity.

FIGURE 69 Short SPVXSTR & Long VST1MT

FIGURE 70 Equal Volatility Combined Strategy

0

50

100

150

200

250


Long VST1MT Short SPVXSTR

0

50

100

150

200

250


Equal Volatility Combined Strategy

Amount of SPVXSTR sold and VST1MT bought are same. . Weights for different strategies are calculated using past 3M realized volatility.

Past performance is not a guarantee of future results.


28 August 2013 19

Source: Barclays Research, OptionMetrics, Bloomberg. Data as of 8/27/13. Past performance is not a guarantee of future results. Start date of the strategies is selected to keep in line with the VIX option inception date.

FIGURE 71 Performance Comparison of Alpha Strategies Since 2007

StrategiesAverage Return

Standard Deviation

Sharpe Ratio

Sortino Ratio

Draw-down

1 Week Return

3 Month Return

SPVXSTR versus SPVXMTR

10.2% 16.6% 0.62 0.88 -31.3% -1.0% 2.5%

Delta Hedged 1M ATM VIX Straddle

7.0% 15.8% 0.45 0.53 -38.0% -0.1% 4.2%

VIX 1M 80%-100% 1x2 Put Spread

4.6% 25.2% 0.18 0.25 -57.3% -10.6% -9.8%

VIX Long 1M ATM Put and Short 2M Same Strike Put

5.2% 20.2% 0.26 0.35 -34.8% -8.9% -14.7%

Long VST1MT Short SPVXSTR

17.5% 36.7% 0.48 0.74 -26.3% -2.5% 3.3%

Leveraged Alpha Dynamic VIX

34.5% 23.9% 1.44 2.11 -21.3% -3.0% 4.3%

Equal Volatility Combined Strategy

8.7% 12.6% 0.70 0.90 -16.2% -3.3% -1.1%


28 August 2013 20

APPENDIX


28 August 2013 21

MARKET INTELLIGENCE

VIX Liquidity Landscape Figure 1: Total Vega Outstanding ($ Millions): We express open interest in vega since that allows us to do a comparison across products.

• For SPX and SPY options, we calculate the total vega across all strikes and maturities by multiplying open interest by vega/contract for each option.

• Vega outstanding of the VIX ETPs is calculated by multiplying the open interest of VIX ETPs listed in the Figure 14 by their respective vega/share.

• The vega/share of the ETPs on day t is calculated first by multiplying the price of the ETPs with their participation rates with respect to the respective indices (SPVXSTR and SPVXMTR). This product is then divided by the constant maturity VIX future to get the vega per share of each ETP. The participation rate, price of the ETPs and constant maturity VIX futures of day t-1 is used to calculate the value on day t. We emphasize that this is not a constant and changes over time.

• In the case of VIX futures, vega outstanding equals the open interest of VIX futures times 1000 as the multiplier of each VIX future is 1000 and it has one vega/share.

• Since the delta of VIX options is the sensitivity of the options with respect to VIX futures, vega outstanding of VIX options is simply the sum of the absolute value of delta times open interest of all listed VIX options.

Figure 2: Total Vega Volume ($ Millions): Vega volume of VIX ETPs, VIX futures, and VIX options is defined in a similar manner as in Figure 1 with the trading volumes replacing the open interest.

VIX Liquidity The first VIX ETPs introduced were unleveraged, which meant that they would provide the performance of the underlying index where the index does not change exposure based on market conditions. While the space was initially dominated by unleveraged products, we see that leveraged products (which provide a constant daily long or short leveraged exposure) and dynamic products (which actively shift the magnitude, tenor and direction of their exposure to VIX futures depending on market conditions) are now a significant part of the product mix. Hence for our analysis we divide the ETPs into the three types described above as unleveraged, leveraged and dynamic.

We also look at a metric called “flow”. The flow differs from Vega AUM due to the significant decline in the value of most ETPs over time. We define the daily “flow” as follows: we first calculate the daily dollar flow for each product as the change in number of shares times the current price of the product and then convert this dollar flow into the equivalent number of VIX futures. We then calculate the cumulative flows based on the daily flow series.

These ETPs can also be divided by the part of the VIX futures curve they operate in. Most of the ETPs are concentrated in the short term region of the VIX futures curve and we label its tenor as Short. Those ETPs which have exposure to the mid-term futures are labelled as Mid. There are also ETPs which have exposure all across the VIX futures curve and are labelled as Mixed.

Figure 3: Current Gross ETP AUM by Type: We calculated the Vega outstanding split by type and displayed the current snapshot in Figure 3.


28 August 2013 22

Figure 4: Current Gross ETP AUM by Tenor ($Millions): This chart shows the current Gross ETP AUM across short and mid tenors split further by long, short and short interest vega exposure.

Figure 5: Gross ETP AUM by Type ($Millions): We show the Vega outstanding of ETPs split by type and displayed the time series.

Figure 6: Gross ETP Flow by Type ($Millions): For each type of ETP, we calculated the cumulative Vega “Flow” into the ETP and displayed the time series. Figure 6 simply plots the cumulative value of this daily flow for the three types of products.

Figure 7: Vega Exposure split by Long, Short exposure and Short Interest ($ Millions): We divide the vega exposure in ETPs coming from 3 categories - Long, Short and Short Interest. For each type of exposure we calculated the Vega outstanding and displayed the time series.

Figure 8: Leveraged ETP impact graph: For levered ETPs like TVIX or inverse ETPs like XIV, daily rebalancing is required to keep the leverage ratio constant. We explore the impact of the rebalancing flows of the leveraged ETPs in the publications Index Volatility Weekly - Recent Increase in Vega of VIX ETPs and Update: Index Volatility Weekly - Implications of Recent Dynamics of TVIX.

The number of shares that need to be traded to rebalance the hedge is given by:

( ) Indext

ETFt

PP

rrmm ⎟

⎠⎞

⎜⎝⎛

+−

11

Here “m” is the leverage ratio and “r” is the return of the ETP.

We also show the 22 Day average volume of the first 2 month VIX futures.

VIX Futures and Options Liquidity All the VIX ETPs are portfolios of VIX futures; therefore, the exact hedge for ETP providers would be to buy or sell the equivalent number of VIX futures. It is thus interesting to compare the trends in VIX ETP AUM with the open interest in VIX futures.

Figure 9: VIX futures versus VIX short term ETP vega: Figure 9 plots the open interest in short term (1st and 2nd) VIX futures and the gross ETP vega in ETPs having exposure to short term futures through time along with their ratio.

Figure 10: VIX futures versus VIX mid-term ETP vega: Figure 10 plots the open interest in mid-term (3rd through 7th) VIX futures and the gross ETP vega in ETPs having exposure to mid-term futures through time along with their ratio.

Figure 11: VIX futures and SPX Option Vega Open Interest: Figure 11 compares the trends in open interest in VIX futures and listed SPX options.

Figure 12: VIX and SPVXSTR Option Vega Open Interest: To monitor the liquidity of VIX and options linked to the SPVXSTR index, Open Interest in Vega ($millions) are shown in the figure. Also shown is the ratio of Put/ Call Open Interest (in contract terms) ratio. Open interest in the figure is the aggregate open interest across all the maturities.

Figure 13: VIX and SPVXSTR Option Vega Volume: To monitor the liquidity of VIX and SPVXSTR options, volume in Vega ($millions) are shown in the figure. Also shown is the ratio of Put/ Call Open volume (in contract terms) ratio.

https://live.barcap.com/go/publications/content?contentPubID=FC1794896�





28 August 2013 23

VIX ETPs Comparison Figure 14: Comparison of various VIX ETPs: In this table, we show the current market capitalization and the daily volume across the VIX ETPs vega terms. The list of the ETPs is sorted by the vega outstanding. We show the gross vega split long vega and short vega exposure, and the rolling 1-month flows in the past month for the ETPs. We also include the short interest of ETPs which are long VIX futures.

We also calculate and display the vega per share. The vega per share of the ETPs on day t is calculated first by multiplying the price of the ETPs with their participation rates with respect to the respective indices (SPVXSTR and SPVXMTR). This product is then divided by the constant maturity VIX future to get the vega per share of each ETP. For ETPs which are of dynamic type, we break the index for the ETP into its component indices and its allocations. We calculate the vega of the component indices in the manner described before and then aggregate the vega to get the vega per share.

We also show the impact of the VIX ETPs on the futures markets with daily rebalancing flows. The SPVXSTR and SPVXMTR indices, which are linked to a bulk of the move in the VIX ETP space, are a portfolio of VIX futures that have to be rolled daily to maintain a constant maturity exposure. The SPVXSTR index starts entirely in the first month VIX futures contract on each monthly rebalance date. An equal fraction of the position is rolled to the second month contract each business day, so that by the subsequent rebalance date (corresponding to the Tuesday before the following month’s VIX expiration), only the second month is held. At that point, the second month becomes the front month and the process commences again. The selling front month futures and buying second month futures constitutes the daily rebalancing flow. This flow does not include the flow due to leveraged ETF rebalancing as explained for Figure 8.

VSTOXX Liquidity Landscape Figure 15: Total Vega Outstanding ($ Millions): We express open interest in vega since that allows us to do a comparison across products.

• For SX5E options, we calculate the total vega across all strikes and maturities by multiplying open interest by vega/contract for each option.

• In the case of V2X futures, vega outstanding equals the open interest of V2X futures times 100 as the multiplier of each V2X future is 100 and it has one vega/share.

• Since the delta of V2X options is the sensitivity of the options with respect to V2X futures, vega outstanding of V2X options is simply the sum of the absolute value of delta times open interest of all listed V2X options.

Figure 16: Total Vega Volume ($ Millions): Vega volume of SX5E options, V2X futures and V2X options is defined in a similar manner as in Figure 1 with the trading volumes replacing the open interest.

Figure 17: V2X Option Open Interest: To monitor the liquidity of V2X, Open Interest in number of contracts is shown in the figure. Also shown is the ratio (Put- Call Open Interest)/(Put + Call Open Interest) where a positive ratio means that the Put OI is higher than the Call OI and vice versa. Open interest in the figure is the aggregate open interest across all the maturities.

Figure 18: V2X Option Volume: To monitor the liquidity of V2X, volume in number of Contracts is shown in the figure. Also shown is the ratio (Put- Call Volume)/(Put + Call Open Volume) where a positive ratio means that the Put Volume is higher than the Call Volume and vice versa.


28 August 2013 24

VIX ETPs and Futures Figure 19: Performance of SPVXMTR, SPVXMSTR, 1M and 5M Constant Maturity Futures: Price performance of SPVXSTR and SPVMTR is plotted for a trailing one-year period. Prices are normalized to 100 at the start date to directly compare the performance of the two ETPs. Along with that, we plot the 1M and 5M Constant Maturity VIX futures. The constant maturity futures are calculated by interpolating the VIX futures term structure using an exponential fit. Constant maturity VIX futures help identify the pure price performance of VIX futures, without taking into account the roll cost.

Figure 20: Estimated SPVXSTR and SPVXMTR Monthly Roll Cost: We estimate the roll cost of the SPVXSTR and SPVXMTR using the term structure of constant maturity VIX futures. Assuming that the volatility surface remains constant, roll cost of SPVXSTR can be estimated as 0.5*(VIX 2M future – VIX)/(VIX 1M future). Applying the same assumptions, SPVXMTR roll cost is estimated as 0.5*(VIX 6M future – VIX 4M future)/(VIX 5M future). Constant maturity VIX futures are calculated here by linearly interpolating VIX futures for the given constant maturities.

Figure 21: VIX Futures Term and Open Interest Snapshot: This figure plots the current VIX futures term structure and the term structure from one week back. The figure gives an idea of the movement in the term structure over the past week.

Figure 22: Hedge Ratio of VIX Futures versus SPX: Shorter-dated VIX futures are more sensitive to movements in SPX relative to longer-dated futures. This is reflected in the higher hedge ratio of SPX relative to longer-dated VIX futures, or in other words more longer-dated VIX futures are required to hedge a given SPX position in comparison to shorter-dated futures. Hedge ratio is calculated using simple linear regression of SPX versus VIX futures overlapping weekly returns over the last two years. We also show the hedge ratios for the short-term and mid-term VIX futures indices.

Figure 23: Scatter Plot SPVXSP versus SPX: In this figure we show the scatter plot in the overlapping 22 day returns of SPVXSP vs SPX.

Figure 24: Scatter Plot SPVXMP versus SPX: In this figure we show the scatter plot in the overlapping 22 day returns of SPVXMP vs SPX.

VSTOXX ETPs and Futures Figure 25: Performance of VST1MT, VMT5MT, VSTOXX 1M and 5M Constant Maturity: This figure plots VST1MT (linked to short term VSTOXX futures) and VMT5MT (linked to mid-term VSTOXX futures) price performance for a trailing one year period. Prices are normalized to 100 at the start date to compare the performance of the two ETPs in terms of the difference in their carry cost and reactivity. Also we include the VSTOXX 1M and 5M future to give the volatility futures level.

Figure 26: Estimated VST1MT and VMT5MT Monthly Roll Cost: We estimate the roll cost of the VST1MT and VMT5MT in the same way as done for VIX. The formulas are now 0.5*(VSTOXX 2M future – VSTOXX)/(VSTOXX 1M future) for VST1MT and 0.5*(VSTOXX 6M future – VSTOXX 4M future)/ (VSTOXX 5M future) for VMT5MT.

Figure 27: VSTOXX Futures Term and Open Interest Snapshot: We visualize the movement in the term structure over the past week for VSTOXX futures.

Figure 28: Hedge Ratio of VSTOXX Futures versus SX5E: The same reasoning as in Figure 18 is used for this figure.


28 August 2013 25

Figure 29: Scatter Plot VST1MT versus SX5E: In this figure we show the scatter plot in the overlapping 22 day returns of VST1MT vs SX5E.

Figure 30: Scatter Plot VMT5MT versus SX5E: In this figure we show the scatter plot in the overlapping 22 day returns of VMT5MT vs SX5E.

VIX and VIX ETP Options Figure 31: VIX and SPVXSTR 2M ATM Volatility: To calculate the VIX volatility surface, we first note that the underlying for each VIX option with different maturity is a different VIX future. We calculate the implied volatility for each option contract with the underlying VIX future having the same maturity as the expiry of the option contract. We only keep options with absolute delta between 5 and 75 to remove points where implied volatility calculation is very sensitive to bid ask spreads. We then use a simple quadratic regression for each option expiry to identify and correct other outliers. A localized kernel regression method is employed over the corrected data to parameterize the implied volatility for given standard deltas and expiries. Localized kernel regression method helps us preserve the shape of the implied volatility surface without placing simplified constraints about the shape of the surface. VIX ATM volatility is defined as the implied volatility of 50 delta option.

Unfortunately, the calculation of VIX realized volatility is complicated by the fact that the underlying VIX future systematically becomes more volatile with decreasing time to maturity. To overcome the problem, we use the methodology for calculating the VIX 2M RV described our publication VIX Derivatives: A Poor Practitioner’s Model. To calculate the 2M RV on a given day, we calculate a time series of theoretical VIX futures expiring on the same day going back two months in history. The VIX futures are calculated by using an exponential fit of the actual VIX contract prices. The realized volatility can now be calculated from these theoretical VIX futures values and can be directly compared with the 2M implied volatility.

As discussed previously in VIX Derivatives: A Poor Practitioner’s Model, VIX ATM implied volatility very closely tracks SPX skew and VIX realized volatility. An intuitive explanation for this effect comes from using a simple sticky strike model. We can provide (slightly) more rigour by using an explicit stochastic volatility model. Perhaps the simplest stochastic volatility model is the SABR model where we can write:

1 2 1 2; ;t t t t t t t t tdS S dZ d dZ dZ dZσ σ νσ ρ= = ⟨ ⟩ =

Thus, as opposed to the Black-Scholes model, the volatilityσ is also assumed to be

stochastic and driven by another random process 2dZ . It can be shown that this model

generates a skew in the implied volatilities and can be approximated as:

( ) logimp oo

KKS

σ σ ρυ⎛ ⎞

= − ⎜ ⎟⎝ ⎠

Thus the skew (defined as the coefficient of the log strike) is directly proportional to the volatility of instantaneous volatility. In this simple model VIX volatility will also be proportional to υ and is thus proportional to the SPX skew. As an aside, note that, if the correlation ρ is -1, then we immediately see that this is equivalent to a sticky strike model.

The above model is too simplistic in that it ignores the mean reversion of volatility which leads to a declining term structure of VIX volatility.

In this figure, we use the same linear model for VIX Implied volatility model as used in the publication to calculate the theoretical VIX ATM implied volatility. The parameters for the




28 August 2013 26

linear model are kept same as the ones used in the publication which were calculated from fitting the data between June 2006 and May 2011.

Figure 32: VIX and SPVXSTR 2M ATM Volatility and Realized Volatility Ratio: We plot the ratio of the 2M ATM Implied Volatility and realized volatility ratio as an indicator of the implied realized premium.

Figure 33: VIX and SPVXSTR 2M Skew: We calculate VIX and SPVXSTR skew as: (25D Call Vol)/(50D Call Vol) - (25D Put Vol)/(50D Put Vol). VIX Skew has a strong relation with the value of the underlying VIX future. In the publication VIX Derivatives: A Poor Practitioner’s Model, we have shown how a reasonable proxy for VIX skew can be created using a linear model with the independent variable being the level of the underlying VIX future. In this figure, we use the same model to calculate the theoretical VIX skew. The parameters for the linear model are kept same as the ones used in the publication which were calculated from fitting the data between June 2006 and May 2011.

Figure 34: VIX and SPVXSTR 2M Skew Snapshot: Negative correlation between VIX futures and SPX returns implies that the sign of VIX skew is flipped with respect to SPX skew. In the figure we plot implied volatility of standardized VIX options using the volatility surface to show the skew of 1M and 2M VIX options. Skew here is defined as (25D Put Vol)/(50D Put Vol) - (25D Call Vol)/(50D Call Vol).

Figure 35: VIX and SPVXSTR ATM Volatility and Open Interest Term Structure Snapshot: VIX and SPVXSTR ATM volatility is the implied volatility of 50 delta option for the listed expiries. The figure compares the latest term structure of volatility of VIX with the term structure one week back.

Figure 36: VIX 3M/1M ATM Volatility Ratio: Longer-dated VIX futures are less volatile than shorter-dated futures. As the VIX futures approach expiry, their volatility increases leading to a downward sloping term structure of VIX volatility. The figure shows the ratio of 3M ATM VIX Volatility and 1M ATM VIX Volatility.

VIX Expiration Snapshot Figure 37: VIX Option Expiring Open Interest distribution By Strike ($MM): In this figure, we show the aggregated open interest notional by strike for the nearest option chain.

Figure 38: VIX Dollar Gamma distribution by strike ($MM): In this chart we calculate the Dollar Gamma for the VIX options, defined as the Change in Dollar Delta for 1% move in the underlying and this is done for those options with upcoming expiration.

Figure 39: Historical Open Interest for most liquid strikes ($MM): In this chart we show the historical open interest in time series format for the upcoming expiration.

Figure 40: Open Interest distribution by Expirations ($MM): We display the open interest in each expiry but in dollar notion terms for VIX options.

VSTOXX Options Figure 41: VSTOXX 2M ATM Volatility: To calculate the VSTOXX volatility surface, we first note that the underlying for each V2X option with different maturity is a different V2X future. We calculate the implied volatility for each option contract with the underlying V2X future having the same maturity as the expiry of the option contract. We only keep options with absolute delta between 5 and 75 to remove points where implied volatility calculation is very sensitive to bid ask spreads. We then use a simple quadratic regression for each option expiry to identify and correct other outliers. A localized kernel regression method is employed over the corrected data to parameterize the implied volatility for given




28 August 2013 27

standard deltas and expiries. Localized kernel regression method helps us preserve the shape of the implied volatility surface without placing simplified constraints about the shape of the surface. V2X ATM volatility is defined as the implied volatility of 50 delta option.

Unfortunately, the calculation of V2X realized volatility is complicated by the fact that the underlying V2X future systematically becomes more volatile with decreasing time to maturity. To overcome the problem, we use the methodology for calculating the V2X 2M RV described our publication Understanding the VSTOXX volatility surface, To calculate the 2M RV on a given day, we calculate a time series of theoretical V2X futures expiring on the same day going back two months in history. The V2X futures are calculated by using an exponential fit of the actual VIX contract prices. The realized volatility can now be calculated from these theoretical V2X futures values and can be directly compared with the 2M implied volatility.

As discussed previously in Understanding the VSTOXX volatility surface, V2X ATM implied volatility very closely tracks SX5E skew and V2X realized volatility. An intuitive explanation for this effect comes from using a simple sticky strike model. We can provide (slightly) more rigour by using an explicit stochastic volatility model. Perhaps the simplest stochastic volatility model is the SABR model where we can write:

1 2 1 2; ;t t t t t t t t tdS S dZ d dZ dZ dZσ σ νσ ρ= = ⟨ ⟩ =

Thus, as opposed to the Black-Scholes model, the volatilityσ is also assumed to be

stochastic and driven by another random process 2dZ . It can be shown that this model

generates a skew in the implied volatilities and can be approximated as:

( ) logimp oo

KKS

σ σ ρυ⎛ ⎞

= − ⎜ ⎟⎝ ⎠

Thus the skew (defined as the coefficient of the log strike) is directly proportional to the volatility of instantaneous volatility. In this simple model V2X volatility will also be proportional to υ and is thus proportional to the SX5E skew. As an aside, note that, if the correlation ρ is -1, then we immediately see that this is equivalent to a sticky strike model.

The above model is too simplistic in that it ignores the mean reversion of volatility which leads to a declining term structure of V2X volatility.

In this figure, we use the same linear model for V2X Implied volatility model as used in the publication to calculate the theoretical V2X ATM implied volatility. The parameters for the linear model are kept same as the ones used in the publication which were calculated from fitting the data between March 2010 and February2013.

Figure 42: VIX and SX5E 2M ATM Volatility and Realized Volatility Ratio: We plot the ratio of the 2M ATM Implied Volatility and realized volatility ratio as an indicator of the implied realized premium.

Figure 43: V2X 3M/1M ATM Volatility Ratio: Longer-dated V2X futures are less volatile than shorter-dated futures. As the V2X futures approach expiry, their volatility increases leading to a downward sloping term structure of V2X volatility. The figure shows the ratio of 3M ATM V2X Volatility and 1M ATM V2X Volatility.

Figure 44: V2X 2M Skew: We calculate V2X and SPVXSTR skew as: (25D Call Vol)/(50D Call Vol) - (25D Put Vol)/(50D Put Vol).

Figure 45: V2X/SX5E 2M Implied Volatility Ratio: We show the ratio of V2X 50 Delta volatility and SX5E 2M 50 Delta implied volatility in this figure.




28 August 2013 28

Figure 46: V2X Option Expiring Open Interest distribution By Strike ($MM): In this figure, we show the aggregated open interest notional by strike for the nearest option chain.

VIX and VSTOXX – Relative Value Figure 47: VIX and VSTOXX Futures Term: In this figure, we show the current futures term structure for VIX and VSTOXX. Apart from the current levels, the chart also depicts the % difference of two adjacent futures maturities to quantify the slope of the term structure in US vs. Europe.

Figure 48: VIX versus VSTOXX constant maturity 1M and 5M futures: The chart depicts the spread between the constant maturity first month and fifth month futures between the VIX and the VSTOXX. This allows us to compare the cross regional spread more consistently over time.

Figure 49: Estimated SPVXSTR and VST1MT Monthly Roll Cost: In this figure, we compare the roll cost for short term futures index in US and in Europe as described in Figure 16. As discussed in our earlier publication, Special Report: Dynamic VSTOXX Strategy - Attractive balance between convexity and carry – March 6, 2012, the roll cost for SPVXSTR is typically much higher than that on the VST1MT.

Figure 50: Estimated SPVXMTR and VMT5MT Monthly Roll Cost: In this figure, we compare the roll cost for midterm futures index in US and in Europe as described in Figure 16. As discussed in our earlier publication, Special Report: Dynamic VSTOXX Strategy - Attractive balance between convexity and carry – March 6, 2012, the roll cost for SPVXMTR is similar to that on the VMT5MT as opposed to short term indices where the difference is much more significant.

Figure 51: VIX and VSTOXX 2M Implied Volatility: In this figure we show the time series of VIX and V2X 2 month 50 Delta Implied Volatility.

Figure 52: VIX and VSTOXX 2M IV/RV Ratio: In this figure we compare the time series of VIX 2M 50 Delta Implied Volatility / 2M Realized Volatility and V2X 2M 50 Delta Implied Volatility/2M Realized Volatility.

https://live.barcap.com/go/publications/email?ADH1799511&m=1334210434427&w=TQCBwDrg-FTet1FICEN3i-pSAhQBYvauiGlHDlaV_ZiKqnVt0QTv7tYFrZHyt1yE�





28 August 2013 29

US HEDGING STRATEGIES: METHODOLOGY

In this section, we compare different hedging strategies using VIX options and VIX-based ETPs. We also show rolling 12M SPX Put Spread Collar hedge series.

All of the considered strategies are rebalanced monthly at VIX futures expiration dates. Performance of the strategies is calculated assuming that the hedges are initiated using a monthly total return swap with money borrowed/lent at the daily Fed funds rate to initiate the hedge. All positions are marked to market daily and dividends are assumed to be reinvested continuously. The strategies show the performance of the pure hedge series. So the SPVXSP hedge series is different from the SPVXSP portfolio performance as the hedge series is sized on each rebalance date. If the SPVXSP hedge increases in value on the next rebalance date we buy more SPVXSP shares so that the sizing remains same with respect to the new notional.

We use SPVXSP and SPVXMP (VIX short-term and medium-term future indices) for calculating the performance of the VIX futures based indices. Performance of the strategies is shown from January 2007. A brief description of the strategies and the sizing used is described in the following section.

Sizing the Hedges The most challenging issue in comparing different hedges, particularly when using volatility-based products as hedges, is how to size them. We adopt a more pragmatic risk budget approach towards sizing our hedges. We use 75% monthly Value at Risk (VaR) as the estimate for cost and size for a risk budget of 5% per annum. In other words, we size the different hedges such that the 75% Value at Risk of each standalone hedge (i.e. without the equity overlay) is less than 5% per annum. We use the transaction cost of 5 cent bid ask for VIX options and VIX futures and 4bp bid ask spread for SPX options.

Performance of the strategies is shown from January 2007.

SPX Hedged with VIX ETPs We consider three hedging strategies using VIX ETPs:

SPVXMP Hedge (Figure 53) In our previous publications, we have discussed the advantages of using SPVXMP as an equity hedge. While it works as a hedge during highly volatile environments, it has also carried well during quiet periods, making it a less costly hedge to carry.

In terms of risk-adjusted returns, SPVXMP hedge has performed better than most of the hedges considered since 2007. It has higher returns and lower standard deviation relative to a pure equity portfolio.

SPVXSP Hedge (Figure 53) SPVXSP tracks the shorter-dated VIX futures in contrast to SPVXMP which tracks longer-dated futures. Historically SPVXSP has had a higher cost to carry during quiet periods, which makes it a more expensive hedge. In return, it has a higher convexity during volatile periods.

In terms of performance, while SPVXSP does reduce draw-down during market declines, it also loses value much faster during more normal times. Consequently, it has worse risk-adjusted ratios than a SPVXMP hedge.


28 August 2013 30

Dynamic VIX (SPDIVXP), BDVIX and LDVIX Hedge (Figure 54) Since SPVXSP has a much higher carry cost, a popular alpha strategy is to go long SPVXMP and go short SPVXSP. This can also be used as a hedge for SPX by maintaining a high SPVXMP:SPVXSP ratio such that it retains a long volatility characteristic and the short SPVXSP leg could limit losses during quiet periods with high roll cost. But, this strategy does not perform so well during the risk flares. Ideally, it would be better to switch from short SPVXSP position to long SPVXSP position as signs of stress begin to emerge.

In this figure, we show performance of SPDVIXP and in terms of performance, SPDVIXP has worked as a hedge and has outperformed a long SPX position. The SPDVIXP hedge outperformed the static short-SPVXSP long SPVXMP strategy during the third quarter of 2008.

A more conservative strategy would be to always have a long volatility exposure without any short component and we label this as the “Long-only Dynamic VIX strategy” (LDVIX).This is introduced in Global Volatility Outlook 2013: A year of fiscal reckoning. This is a quintessential hedging strategy and it avoids some of the risks inherent in SPDVIXP (which leads to higher costs) and could be appropriate for more conservative investors. The strategy has allocations of 25% and 50% to SPVXMP in the first and second buckets, respectively. The remaining three buckets are unchanged from SPDVIXP.

An alternative strategy is to have no allocation to VIX futures in the first two buckets (strong contango) and have the same allocations as SPDVIXP in the remaining three buckets. We term this the “Backwardation Dynamic VIX Strategy” BDVIX strategy. The idea is to have a hedge that provides long volatility exposure during high volatility periods but has no volatility exposure during quiet periods. The main attraction of this strategy is the staircase kind of payoff profile it has provided historically, which can be attractive for some investors. However, this involves a strong reliance on the signal since the strategy has zero exposure to volatility unless the signal indicates that it needs to. This strategy could work well for selloffs, which last a few days but would miss out on most small equity declines.

The sizing for all strategies is the same to the Dynamic VIX future (SPDVIXP).

VIX Options Hedges We consider three VIX options-based hedging strategies. We had discussed VIX options hedging strategies in detail in our publication: Index Volatility Weekly - VIX Options Trading Strategies: A Deep Dive – April 19, 2010.

VIX 3M Calls Hedge (Figure 55) VIX Calls are a natural hedge for the equity due to the negative correlation of the VIX with the equity prices and the convexity it offers. Each month we buy three-month 20% OTM call options. The options are rolled after one month when the options have two months left to expiration.

The premium required to buy the option is borrowed each month at the Fed funds rate. Equity position is adjusted at each option expiration date according to the gain or loss from the option position and the interest incurred on the borrowing.

While VIX call options lose money in most months, they tend to work well in instances where equity markets experience large negative returns. In that respect, they are much more of a tail hedge, which is why we prefer 20% OTM call options in comparison to ATM call options. It is also cheaper to carry longer-dated option since the implied volatility of VIX options has usually been the richest in the shorter end. In terms of performance, they outperform long equity but underperform some other VIX option hedges that try to reduce the cost of the hedge by selling some other option in addition to the long call leg.


https://live.barcap.com/go/research/content?contentPubID=FC1588251�



28 August 2013 31

VIX 1M Calls + Short VIX 1M DH Straddle Hedge (Figure 55) One way to reduce the cost of out-of-the-money VIX 1M calls is by shorting VIX delta hedged straddles. The Short VIX Delta-hedged straddle strategy has worked historically as it monetizes the rich implied realized volatility risk premium present in VIX options. In our publication Index Volatility Weekly - Cheapening the cost of VIX Options Based Hedges, we first discussed this strategy where we short VIX ATM straddles and buy 120% VIX 1M calls. Delta-hedged straddles have path dependence; as the underlying moves away from the ATM strike, the gamma reduces and the P/L from the straddle becomes progressively lower. This is an important consideration in combining the short VIX delta-hedged straddle with long OTM VIX call since a large move up in VIX can make the P/L contribution of the VIX straddle relatively less important and the hedge effectively becomes long VIX futures (since the VIX call option delta also becomes close to one).

The primary risk to the performance of the hedge would involve VIX futures having a high realized volatility while remaining close to the ATM strike. Another risk would be a large decline in the VIX future close to the VIX options expiration. Given that implied volatility of VIX options usually trades at a significant premium to realized volatility, there is a buffer for the delta-hedged straddle even if realized volatility increases.

In our opinion, the aim of the VIX straddle overlay should be to make the VIX OTM call hedge effectively costless during rising markets while working as a hedge when volatility sharply rises.

VIX Risk Reversals & Static Short SPVXSP Long SPVXMP Hedge (Figure 56) VIX call options are expensive due to the high implied realized volatility premium of VIX options. Cost of buying the call options can be reduced by selling OTM put options on VIX. Similar to the call hedge, we buy one-month 20% OTM call options to hedge SPX but also sell one-month 20% OTM put options on VIX to reduce the cost of the calls. The premium paid (or received) from buying and selling the options is borrowed (or deposited) at the Fed funds rate.

Historically, VIX has not had many spikes to the downside and even though OTM puts are cheaper than OTM calls, they are still a good sell, in our opinion. The risk of being short these puts is mitigated by the convexity of VIX to SPX, which ensures that the downside to volatility is capped. In terms of performance, this works the best among the VIX options-based equity hedges. The static short-SPVXSP long-SPVXMP strategy is also displayed. The notional for the short SPVXSP leg is kept to be one fourth of the SPVXMTR leg.

Short VIX 1x2 Calls Spreads Hedge (Figure 57) The reason we want to go long far OTM calls is that most spikes in volatility are very pronounced, which makes the far right tail a desirable part of the VIX distribution. In contrast, the intermediate part of the distribution is usually overpriced and a good sell. In other words, VIX usually goes up a lot or not at all.

Selling a VIX 1x2 call spread allows for going long a far OTM call which is funded by selling a nearer call spread. The hedge would cost much less than a long VIX call and would work during large volatility spikes. The risk to this strategy involves a positive drift for VIX.

We implement this hedge in the three-month space which is cheaper to carry than the front month. Each month we sell three-month 120-150 VIX 1x2 call spread and roll it after one month to a new three-month option. The strategy does not cost much to carry but also provides less upside during crisis situations.



28 August 2013 32

VIX Calendar Strangle Hedge (Figure 57) The VIX calendar strangle strategy, which funds a long VIX call position by collecting the roll yield in VIX futures by being long VIX puts, has gained popularity. It performed well in 2012 due to the high returns generated by the long put leg ( Global Volatility Outlook 2013: A year of fiscal reckoning).

The VIX calendar strangle strategy seeks to fund the long VIX call position by using a long position in VIX puts to monetize the higher roll yield in shorter-dated futures. Thus the puts are deliberately chosen to be shorter maturity since that is where the roll yield is the highest. On the other hand, as we have discussed before, it is more optimal to use longer-dated VIX calls, which have a lower volatility risk premium and lower roll cost.

While VIX call options show higher convexity and less basis risk, the calendar strangle strategy can have negative returns for declines in SPX. The strategy works for cases where SPX increases significantly, as volatility declines sharply in these cases and long VIX puts benefits the strategy.

We select a VIX 1M-3M strangle with the strikes of the puts and calls taken as the closest strike to 95% and 120% of the underlying VIX futures, respectively. While the usual implementation uses equal notional for calls and puts, we also show the performance of a strategy where the notional of the put is half that of the calls. This variation puts less emphasis on earning the carry from the term structure premium.

Rolling SPX Put Spread Collar Hedge We also show the performance of SPX rolling put spread collar hedge series in Figure 48, which is one of the best performing conventional options-based strategies. The strategy involves buying a 12M 15-40 Delta SPX put spread each month with 1/12 the notional of the hedge position and selling 1M 25 Delta SPX Call and keeping the options to expiration. While this hedge outperforms a stand-alone long equity position, it underperforms most of the VIX ETPs and VIX options-based hedges.




28 August 2013 33

EU HEDGING STRATEGIES: METHODOLOGY

In this section, we compare different hedging strategies for an SX5E portfolio. The hedging instruments include vanilla SX5E puts and put spread collars and VSTOXX-based ETPs. We add to this the dynamic VSTOXX strategy that we developed and discussed in the report: Special Report: Dynamic VSTOXX Strategy - Attractive balance between convexity and carry, 6 Mar 2012.

All of the considered strategies are rebalanced monthly at SX5E options expiration dates. To calculate the performance of the strategies we assume that the hedges are initiated using a monthly total return swap with money borrowed/lent at the daily borrow rate (EONIA) to initiate the hedge. We adjust the equity exposure (SX5E) in our portfolio each month at rebalance dates according to the gain or loss on the hedge.

In addition to the time series of SX5E + hedge, we also show the excess return series for the hedge itself, Figures 50 and 52. This helps us isolate the pure hedge performance from the underlying equity performance. We use VST1ME and VMT5ME (V2X short-term and medium-term future indices) for calculating the performance of the V2X futures based indices.

Sizing the Hedges The most challenging issue in comparing different hedges, particularly when using volatility-based products as hedges, is how to size them. We adopt a more pragmatic risk budget approach towards sizing our hedges. We use 75% monthly Value at Risk (VaR) as the estimate for cost and size for a risk budget of 5% per annum. In other words, we size the different hedges such that the 75% Value at Risk of each standalone hedge (i.e. without the equity overlay) is less than 5% per annum. The resulting allocation is shown in Figure 64.

Performance of the strategies is shown from January 2007.

SX5E Hedged with VSTOXX Based Indices We consider three hedging strategies for SX5E using VSTOXX based indices:

SX5E Hedged with VST1MT (Figures 60, 61) In our previous publications (please see: Volatility investing with the VSTOXX, 1 Mar 2010), we highlighted the diversification benefits of adding volatility to an equity portfolio due to its high negative correlation with equity markets and high convexity in negative equity market returns. VST1ME, which tracks the shorter-dated VSTOXX futures, has had a higher cost to carry during quiet periods, which makes it a more expensive hedge. In return, it has a higher convexity during volatile periods. Due to its high cost of carry we allocate only €3.5 to VST1ME for €100 equity portfolio, under our risk budget approach as described in the section ‘Sizing the Hedges’.

SX5E Hedged with VMT5MT (Figures 60, 61) We have discussed the advantages of using VMT5ME (medium dated forward volatility) as an equity hedge in our previous publications (please see: Index Derivatives Weekly: Simulating a mid-term VSTOXX futures index, 4 Oct 2010). While it works as a hedge during highly volatile environments, it has also carried well during quiet periods, making it a less costly hedge to carry. This translates into a slightly higher allocation for VMT5ME (€7.5 for €100 of equity) compared to VST1ME.

VMT5ME hedge has provided a better risk return profile versus a pure equity portfolio for the period considered. It has higher returns and lower standard deviation relative to a pure equity portfolio.



https://live.barcap.com/go/publications/email?ADH1573599&m=1334210434427&w=6gdjg4u7bVWIkKk97ytZXkob0TdowPRcGUAJVt79QuWSBp8C9iiFCLvyMysQCFnX�

https://live.barcap.com/go/publications/email?ADH1639836&m=1334210434427&w=ra4ER5mjOqtZtqSNaXZEEGjpx3ulPnnzJXFQ0KifSbVNRLuwoiEri5Z323zSZI1Q�

https://live.barcap.com/go/publications/email?ADH1639836&m=1334210434427&w=ra4ER5mjOqtZtqSNaXZEEGjpx3ulPnnzJXFQ0KifSbVNRLuwoiEri5Z323zSZI1Q�


28 August 2013 34

SX5E Hedged with Dynamic VSTOXX (Figures 60, 61) Shorter-dated VSTOXX futures/indices have a much higher carry cost than the medium-term futures. This, in turn, is consistent with the fact that the former are much more convex and reactive to equity returns. A consequence is that short-term futures are expensive to carry over longer horizons and more suitable for short-term tactical trading, in our view. On the other hand, despite their lesser convexity, medium-term futures and associated indices are much more efficient as a buy-and-hold portfolio hedge.

Ideally, one would like to benefit from the convexity of a short-term volatility overlay during an equity sell-off, while generating more constrained carry cost during benign market periods. In our publication (please see: Special Report: Dynamic VSTOXX Strategy - Attractive balance between convexity and carry, 6 Mar 2012), we highlighted that a good way to manage this trade-off between convexity and carry is to dynamically allocate to short- and long-term volatility based on implied volatility term structure slope. The strategy switches from short VST1ME position to long VMT5ME position as signs of stress begin to emerge while maintaining long exposure to VMT5ME throughout. The strategy design attempts to maintain a long volatility bias across different periods in time.

The better carry of the strategy can also be seen from the relatively higher allocation to dynamic VSTOXX under our risk budget approach (€13.5 for €100 of equity, compared to €3.5 and €7.5 for VST1ME and VMT5ME respectively). The dynamic VSTOXX hedge provides a much superior risk return profile versus other hedges.

SX5E Hedged with Vanilla Options As a comparison to hedges based on VSTOXX ETPs, we also show a couple of plain vanilla option-based hedges – the simplest plain vanilla put hedge and possibly one of the best-performing conventional option-based strategies; i.e. put spread collar.

SX5E Hedged with a Put (Figures 62, 63) We show the performance of SX5E hedged with a 1 month put on SX5E, which is one of the most direct conventional options-based hedge. The strategy involves buying a 1 month 25-delta SX5E put each month holding it to expiration. The strategy provides high convexity during sell-offs but the hedge performance indicates that the options are usually priced very expensive, which causes a high drag on portfolio performance during benign market conditions. Therefore, systematically buying put options is an inefficient hedge, in our opinion.

SX5E Hedged with a Rolling Put Spread Collar (Figures 62, 63) We also show the performance of SX5E hedged with a rolling put spread collar, which is arguably one of the best performing conventional options-based strategies. The strategy involves buying a 12 month 40-delta SX5E put financed by simultaneously selling a 12 month 15-delta put and a 25-delta call each month with 1/12 the notional of the equity position and keeping the options to expiration. The hedge carries much better than an outright put but at expense of capping the upside and at expense of not providing the tail risk protection.




28 August 2013 35

ALPHA STRATEGIES: METHODOLOGY

In the previous section, we discussed how VIX options and VIX ETPs work as equity hedges. While VIX products are attractive hedges, they also offer alpha opportunities due to the usually steep contango in VIX futures and the high volatility of VIX options. All four strategies discussed below attempt to take advantage of one or more of these characteristics.

We show the performance of all of these strategies in the form of an index with daily prices. Since these are all different variations of a carry strategy, they are likely to experience draw-down risk flares. It becomes very important to correctly size these trades. Our aim is to be conservative in putting on risk so that it becomes very unlikely that the index will have outsized draw-downs. We look at the worst returns of each of these strategies over a rolling one-month period and size the trade such that it will roughly require four times the worst return for the index to go to zero. In other words, we are restricting the worst one-month draw-down historically to be around 25% of the index. We use the transaction cost of 5bps bid ask for VIX options and VIX futures and 4bp bid ask spread for SPX options.

Short SPVXSTR Long SPVXMTR In previous publications, we showed how to monetize the VIX term structure premium through SPVXSTR and SPVXMTR. Selling SPVXSTR allows us to capture high roll yields present between the first and second month futures but it suffers from high draw-down when the volatility spikes. Buying SPVXMTR, along with selling SPVXSTR, reduces the draw-down and improves the performance considerably due to the flatter term structure between four- and seven-month VIX futures. We size the trade such that we go long $2 of SPVXMTR for every $1 short of SPVXSTR. The ratio of 2 is close to the average beta of SPVXMTR relative to SPVXSTR over a long history.

The worst loss experienced by the strategy over a rolling one-month period is 50% of the SPVXSTR notional during October 2008. We therefore size the strategy such that we sell $50 of SPVXSTR and buy $100 of SPVXMTR at each rebalancing date for every $100 of equity in the portfolio. This limits the historical maximum draw-down to 25% of our strategy.

The strategy is implemented in the form of a one-month total return swap with the equity being adjusted at the rebalance date (which is the VIX expiration date) according to the P/L of the swap.

The strategy has worked historically. One can think of the short SPVXSTR leg as the alpha part of the trade with SPVXSTR working well as a hedge.

Leveraged Alpha Dynamic VIX (LAVIX) The Dynamic VIX signal could be leveraged to create an alpha strategy by increasing the negative exposure to short-term futures during risk-on phases when the signal indicates negative returns for volatility. We introduced this strategy in our Global Volatility Outlook 2013: A year of Fiscal Reckoning. In the contango buckets, LAVIX is much more aggressive and in the two contango buckets (first two rows), LAVIX has a significant net short volatility exposure with gross notional exceeding 100% while DVIX always remains long volatility. In backwardation buckets (bottom three rows), while LAVIX has an overall long volatility exposure, the long volatility exposure is less than that in DVIX in each of the three buckets. The use of the signal, which has achieved positive jumps in the past, can also lead to negative jumps if it does not correctly position the strategy prior to volatility spikes. The performance has generally been good in times of stress as the signal has in most cases correctly positioned the strategy to be long volatility, but the fact that it has a negative volatility exposure in the first two buckets makes it more susceptible to draw-downs.




28 August 2013 36

Delta Hedged VIX Straddle As discussed in the report Volatility of Volatility Risk Premium – February 16, 2010, a high implied realized premium exists in VIX options. One of the simplest ways to monetize this premium is by selling a VIX one-month delta hedged straddle.

The risk to this trade is that subsequent one-month VIX volatility is much higher than the implied volatility. We estimate that the highest realized volatility for front month VIX futures was ~250% during 2001 when the previous one-month volatility was only 50%. Using an estimate that the VIX implied volatility was 75%, the loss would have been 250 – 75 = 175 volatility points. Assuming our rule of a maximum 25% draw-down, this would translate into a vega of ~0.15 per $100 of equity. As we have shown before, front month VIX options have historically been the most expensive, which makes them good candidates for selling delta hedged straddles.

Each month we sell 0.15 Vega of ATM straddles for $100 of equity. The straddle is delta hedged on a daily basis. The money received from selling the straddles is deposited at the Fed funds rate. The cash level is adjusted at the expiry for the gain or loss from selling the straddles and delta hedging them.

It is interesting to note that the realized volatility of VIX options remains in a narrower band of its average implied volatility relative to SPX, where the divergence can be much greater. This makes selling VIX straddles attractive relative to a similar trade in SPX which also has a lower Sharpe ratio during the same period.

VIX 1x2 Put Spread The usual steep term structure of VIX futures indicates that the VIX future would settle lower if the volatility term structure remains constant. In that regard, buying VIX put options allows for monetizing this roll down the curve. However, as was highlighted in the VIX straddle strategy, VIX options are usually expensive due to their high implied realized volatility premium. VIX 1x2 put spreads are short volatility of volatility and have a low tail risk due to the rarity of large downward moves in VIX. We had discussed the benefits of VIX 1x2 put spreads in our publication Index Volatility Weekly - VIX Options Trading Strategies: A Deep Dive – April 19, 2010.

We select VIX one-month 80-100 1x2 put spread. The strikes were selected by comparing the historical performance of 1x2 put spread of different strikes. The cost of 1x2 put spreads has usually averaged 7% and the maximum loss experienced by the strategy would be limited to ~25% of the notional even in an extreme case of the front VIX future losing 60% of value in one month. In line with our methodology of limiting the one-month draw-down to 25%, we size the trade such that we buy ATM puts on a notional of $100 for every $100 of equity.

Each month, we buy the 1x2 put spreads with the notional of the long leg equalling our equity position. The cash outlay for buying the spread is borrowed at the fed funds rate and the P/L of the strategy at the end of the month is added to the equity position.

VIX Calendar Spread Another way to monetize the steep contango in VIX futures through VIX options is by going short a VIX put calendar spread (buy near-term put and sell longer-dated put). By buying a near-term put and selling a longer-term put of the same strike, we have a short position in near-term VIX future and long position in longer-dated VIX future, which is similar to the SPVXSTR versus SPVXMTR trade. We buy the front month ATM put and sell the second-month put with the same strike.





28 August 2013 37

The trade in many ways is similar to the 1x2 put spread with the cost of buying the front month ATM put being funded by selling the longer maturity put. In the case of 1x2 put spreads, the cost of the ATM put was funded by selling lower strike puts. The risk to the calendar trade is for the term structure to invert with the longer-dated put being in the money while the shorter-dated put expires out of the money. However, the inversion of the term structure is usually associated with an increase in the levels of volatilities, which should result in even the second month options being out of the money.

While it is difficult to estimate the possible losses in the trade, in many ways it is a less risky trade than the 1x2 put spread or the SPVXSTR vs. SPVXMTR trade since an inversion of the surface is likely to be associated with a leg up for volatilities, which should keep the draw-down low. We size the trade similar to the 1x2 put spread with the notional of both front month and second month options equaling the cash held.

Short SPVXSTR Long VST1MT In our research, we have discussed monetizing relatively higher VIX term structure premium through selling SPVXSTR and buying short-term VSTOXX futures index (VST1MT). The strategy benefits from the outperformance of VST1MT over SPVXSTR. This outperformance is more a result of the steeper term structure of SPX volatility relative to SX5E volatility than any divergence in the volatility levels of SPX and SX5E. We size the trade such that we go long $50 of VST1MT and short $50 of SPVXSTR for every $100 of index.

The strategy is implemented in the form of a one-month total-return swap with the equity being adjusted at the rebalance date (which is the VIX expiration date) according to the P/L of the swap.

Historically, the strategy has worked in most of the period from the beginning of 2006, apart from the period between March and December 2007. The underperformance in this period can be explained by the lag in the rise of volatility in Europe relative to the rise in the US.

Ideally, P/L should be calculated using the prices of both the indices at the same time. Since we do not have a long enough history of intraday prices, we use closing levels of the indices to calculate the P/L. However, the calculation of P/L from closing prices leads to spurious noise, resulting in higher standard deviation and lower Sharpe Ratio. In that regard, using weekly returns for calculating the Sharpe Ratio leads to much better performance metrics.

Equal Volatility Combined Strategy As we have discussed before, the correlation between the different VIX alpha strategies is relatively low, with only the VIX 1x2 put spread and the VIX Calendars strategies having a high correlation between them. The low correlation between the strategies indicates that the strategies can be combined to achieve better risk-adjusted returns.

We have used the trailing volatility of the strategies to size them, which leads to lower weights for strategies with high volatility. This method of combining the strategies results in better performance when predictability of correlation is low. The combined strategy is rebalanced every month at the VIX expiration dates to reassign weights to the strategies. At the VIX expiration date, the new weight assigned to each strategy is inversely proportional to the trailing three-month realized volatility of the strategy calculated using weekly overlapping returns.

Historically, the combined strategy has achieved higher risk-adjusted returns and lower draw-downs compared to the other VIX strategies considered.


28 August 2013 38

ANALYST(S) CERTIFICATION(S):

We, Maneesh S. Deshpande, Rohit Bhatia and Christian Kober, hereby certify (1) that the views expressed in this research report accurately reflect our personal views about any or all of the subject securities or issuers referred to in this research report and (2) no part of our compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this research report.

IMPORTANT DISCLOSURES CONTINUED

Barclays Research is a part of the Corporate and Investment Banking division of Barclays Bank PLC and its affiliates (collectively and each individually, "Barclays"). For current important disclosures regarding companies that are the subject of this research report, please send a writtenrequest to: Barclays Research Compliance, 745 Seventh Avenue, 14th Floor, New York, NY 10019 or refer to http://publicresearch.barclays.com or call 212-526-1072.

The analysts responsible for preparing this research report have received compensation based upon various factors including the firm's total revenues, a portion of which is generated by investment banking activities.

Research analysts employed outside the US by affiliates of Barclays Capital Inc. are not registered/qualified as research analysts with FINRA. These analysts may not be associated persons of the member firm and therefore may not be subject to NASD Rule 2711 and incorporated NYSE Rule 472 restrictions on communications with a subject company, public appearances and trading securities held by a research analyst’s account.

Analysts regularly conduct site visits to view the material operations of covered companies, but Barclays policy prohibits them from accepting payment or reimbursement by any covered company of their travel expenses for such visits.

In order to access Barclays Statement regarding Research Dissemination Policies and Procedures, please refer tohttps://live.barcap.com/publiccp/RSR/nyfipubs/disclaimer/disclaimer-research-dissemination.html. In order to access Barclays Research Conflict Management Policy Statement, please refer to: http://group.barclays.com/corporates-and-institutions/research/research-policy.

The Corporate and Investment Banking division of Barclays produces a variety of research products including, but not limited to, fundamental analysis, equity-linked analysis, quantitative analysis, and trade ideas. Recommendations contained in one type of research product may differ from recommendations contained in other types of research products, whether as a result of differing time horizons, methodologies, or otherwise.

Risk Disclosure(s)

Options are not suitable for all investors. Please note that the trade ideas within this research report do not necessarily relate to, and may directlyconflict with, the fundamental ratings applied to Barclays Equity Research. The risks of options trading should be weighed against the potential rewards.

Risks:

Call or put purchasing: The risk of purchasing a call/put is that investors will lose the entire premium paid.

Uncovered call writing: The risk of selling an uncovered call is unlimited and may result in losses significantly greater than the premium received.

Uncovered put writing: The risk of selling an uncovered put is significant and may result in losses significantly greater than the premium received.

Call or put vertical spread purchasing (same expiration month for both options): The basic risk of effecting a long spread transaction is limited to the premium paid when the position is established.

Call or put vertical spread writing/writing calls or puts (usually referred to as uncovered writing, combinations or straddles; same expiration month for both options): The basic risk of effecting a short spread transaction is limited to the difference between the strike prices less the amount received in premiums.

Call or put calendar spread purchasing (different expiration months; short must expire prior to the long): The basic risk of effecting a long calendar spread transaction is limited to the premium paid when the position is established.

Because of the importance of tax considerations to many options transactions, the investor considering options should consult with his/her tax advisor as to how taxes affect the outcome of contemplated options transactions.

Supporting documents that form the basis of our recommendations are available on request.

The Options Clearing Corporation's report, "Characteristics and Risks of Standardized Options", is available athttp://www.theocc.com/about/publications/character-risks.jsp

Barclays offices involved in the production of equity research:

London

Barclays Bank PLC (Barclays, London)

New York

Barclays Capital Inc. (BCI, New York)

Tokyo

Barclays Securities Japan Limited (BSJL, Tokyo)

São Paulo

Banco Barclays S.A. (BBSA, São Paulo)

Hong Kong


28 August 2013 39

IMPORTANT DISCLOSURES CONTINUED

Barclays Bank PLC, Hong Kong branch (Barclays Bank, Hong Kong)

Toronto

Barclays Capital Canada Inc. (BCCI, Toronto)

Johannesburg

Absa Capital, a division of Absa Bank Limited (Absa Capital, Johannesburg)

Mexico City

Barclays Bank Mexico, S.A. (BBMX, Mexico City)

Taiwan

Barclays Capital Securities Taiwan Limited (BCSTW, Taiwan)

Seoul

Barclays Capital Securities Limited (BCSL, Seoul)

Mumbai

Barclays Securities (India) Private Limited (BSIPL, Mumbai)

Singapore

Barclays Bank PLC, Singapore branch (Barclays Bank, Singapore)

DISCLAIMER:

This publication has been prepared by the Corporate and Investment Banking division of Barclays Bank PLC and/or one or more of its affiliates (collectivelyand each individually, "Barclays"). It has been issued by one or more Barclays legal entities within its Corporate and Investment Banking division as providedbelow. It is provided to our clients for information purposes only, and Barclays makes no express or implied warranties, and expressly disclaims all warrantiesof merchantability or fitness for a particular purpose or use with respect to any data included in this publication. Barclays will not treat unauthorizedrecipients of this report as its clients. Prices shown are indicative and Barclays is not offering to buy or sell or soliciting offers to buy or sell any financialinstrument.

Without limiting any of the foregoing and to the extent permitted by law, in no event shall Barclays, nor any affiliate, nor any of their respective officers,directors, partners, or employees have any liability for (a) any special, punitive, indirect, or consequential damages; or (b) any lost profits, lost revenue, loss ofanticipated savings or loss of opportunity or other financial loss, even if notified of the possibility of such damages, arising from any use of this publication orits contents.

Other than disclosures relating to Barclays, the information contained in this publication has been obtained from sources that Barclays Research believes tobe reliable, but Barclays does not represent or warrant that it is accurate or complete. Barclays is not responsible for, and makes no warranties whatsoever asto, the content of any third-party web site accessed via a hyperlink in this publication and such information is not incorporated by reference.

The views in this publication are those of the author(s) and are subject to change, and Barclays has no obligation to update its opinions or the information inthis publication. The analyst recommendations in this publication reflect solely and exclusively those of the author(s), and such opinions were preparedindependently of any other interests, including those of Barclays and/or its affiliates. This publication does not constitute personal investment advice or takeinto account the individual financial circumstances or objectives of the clients who receive it. The securities discussed herein may not be suitable for allinvestors. Barclays recommends that investors independently evaluate each issuer, security or instrument discussed herein and consult any independentadvisors they believe necessary. The value of and income from any investment may fluctuate from day to day as a result of changes in relevant economicmarkets (including changes in market liquidity). The information herein is not intended to predict actual results, which may differ substantially from thosereflected. Past performance is not necessarily indicative of future results.

This communication is being made available in the UK and Europe primarily to persons who are investment professionals as that term is defined in Article 19of the Financial Services and Markets Act 2000 (Financial Promotion) Order 2005. It is directed at, and therefore should only be relied upon by, persons whohave professional experience in matters relating to investments. The investments to which it relates are available only to such persons and will be enteredinto only with such persons. Barclays Bank PLC is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority andthe Prudential Regulation Authority and is a member of the London Stock Exchange.

The Corporate and Investment Banking division of Barclays undertakes U.S. securities business in the name of its wholly owned subsidiary Barclays CapitalInc., a FINRA and SIPC member. Barclays Capital Inc., a U.S. registered broker/dealer, is distributing this material in the United States and, in connectiontherewith accepts responsibility for its contents. Any U.S. person wishing to effect a transaction in any security discussed herein should do so only bycontacting a representative of Barclays Capital Inc. in the U.S. at 745 Seventh Avenue, New York, New York 10019.

Non-U.S. persons should contact and execute transactions through a Barclays Bank PLC branch or affiliate in their home jurisdiction unless local regulationspermit otherwise.

Barclays Bank PLC, Paris Branch (registered in France under Paris RCS number 381 066 281) is regulated by the Autorité des marchés financiers and theAutorité de contrôle prudentiel. Registered office 34/36 Avenue de Friedland 75008 Paris.

This material is distributed in Canada by Barclays Capital Canada Inc., a registered investment dealer and member of IIROC (www.iiroc.ca).

Subject to the conditions of this publication as set out above, Absa Capital, the Investment Banking Division of Absa Bank Limited, an authorised financialservices provider (Registration No.: 1986/004794/06. Registered Credit Provider Reg No NCRCP7), is distributing this material in South Africa. Absa BankLimited is regulated by the South African Reserve Bank. This publication is not, nor is it intended to be, advice as defined and/or contemplated in the (SouthAfrican) Financial Advisory and Intermediary Services Act, 37 of 2002, or any other financial, investment, trading, tax, legal, accounting, retirement, actuarialor other professional advice or service whatsoever. Any South African person or entity wishing to effect a transaction in any security discussed herein shoulddo so only by contacting a representative of Absa Capital in South Africa, 15 Alice Lane, Sandton, Johannesburg, Gauteng 2196. Absa Capital is an affiliate ofBarclays.

In Japan, foreign exchange research reports are prepared and distributed by Barclays Bank PLC Tokyo Branch. Other research reports are distributed toinstitutional investors in Japan by Barclays Securities Japan Limited. Barclays Securities Japan Limited is a joint-stock company incorporated in Japan withregistered office of 6-10-1 Roppongi, Minato-ku, Tokyo 106-6131, Japan. It is a subsidiary of Barclays Bank PLC and a registered financial instruments firmregulated by the Financial Services Agency of Japan. Registered Number: Kanto Zaimukyokucho (kinsho) No. 143.

Barclays Bank PLC, Hong Kong Branch is distributing this material in Hong Kong as an authorised institution regulated by the Hong Kong MonetaryAuthority. Registered Office: 41/F, Cheung Kong Center, 2 Queen's Road Central, Hong Kong.

This material is issued in Taiwan by Barclays Capital Securities Taiwan Limited. This material on securities not traded in Taiwan is not to be construed as'recommendation' in Taiwan. Barclays Capital Securities Taiwan Limited does not accept orders from clients to trade in such securities. This material maynot be distributed to the public media or used by the public media without prior written consent of Barclays.

This material is distributed in South Korea by Barclays Capital Securities Limited, Seoul Branch.

All equity research material is distributed in India by Barclays Securities (India) Private Limited (SEBI Registration No: INB/INF 231292732 (NSE), INB/INF011292738 (BSE), Registered Office: 208 | Ceejay House | Dr. Annie Besant Road | Shivsagar Estate | Worli | Mumbai - 400 018 | India, Phone: + 91 2267196363). Other research reports are distributed in India by Barclays Bank PLC, India Branch.

Barclays Bank PLC Frankfurt Branch distributes this material in Germany under the supervision of Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin).

This material is distributed in Malaysia by Barclays Capital Markets Malaysia Sdn Bhd.

This material is distributed in Brazil by Banco Barclays S.A.

This material is distributed in Mexico by Barclays Bank Mexico, S.A.

Barclays Bank PLC in the Dubai International Financial Centre (Registered No. 0060) is regulated by the Dubai Financial Services Authority (DFSA). Principalplace of business in the Dubai International Financial Centre: The Gate Village, Building 4, Level 4, PO Box 506504, Dubai, United Arab Emirates. BarclaysBank PLC-DIFC Branch, may only undertake the financial services activities that fall within the scope of its existing DFSA licence. Related financial products orservices are only available to Professional Clients, as defined by the Dubai Financial Services Authority.

Barclays Bank PLC in the UAE is regulated by the Central Bank of the UAE and is licensed to conduct business activities as a branch of a commercial bankincorporated outside the UAE in Dubai (Licence No.: 13/1844/2008, Registered Office: Building No. 6, Burj Dubai Business Hub, Sheikh Zayed Road, DubaiCity) and Abu Dhabi (Licence No.: 13/952/2008, Registered Office: Al Jazira Towers, Hamdan Street, PO Box 2734, Abu Dhabi).

Barclays Bank PLC in the Qatar Financial Centre (Registered No. 00018) is authorised by the Qatar Financial Centre Regulatory Authority (QFCRA). BarclaysBank PLC-QFC Branch may only undertake the regulated activities that fall within the scope of its existing QFCRA licence. Principal place of business in Qatar:Qatar Financial Centre, Office 1002, 10th Floor, QFC Tower, Diplomatic Area, West Bay, PO Box 15891, Doha, Qatar. Related financial products or servicesare only available to Business Customers as defined by the Qatar Financial Centre Regulatory Authority.

This material is distributed in the UAE (including the Dubai International Financial Centre) and Qatar by Barclays Bank PLC.

This material is distributed in Saudi Arabia by Barclays Saudi Arabia ('BSA'). It is not the intention of the publication to be used or deemed asrecommendation, option or advice for any action (s) that may take place in future. Barclays Saudi Arabia is a Closed Joint Stock Company, (CMA License No.09141-37). Registered office Al Faisaliah Tower, Level 18, Riyadh 11311, Kingdom of Saudi Arabia. Authorised and regulated by the Capital MarketAuthority, Commercial Registration Number: 1010283024.

This material is distributed in Russia by OOO Barclays Capital, affiliated company of Barclays Bank PLC, registered and regulated in Russia by the FSFM.Broker License #177-11850-100000; Dealer License #177-11855-010000. Registered address in Russia: 125047 Moscow, 1st Tverskaya-Yamskaya str. 21.

This material is distributed in Singapore by the Singapore branch of Barclays Bank PLC, a bank licensed in Singapore by the Monetary Authority of Singapore.For matters in connection with this report, recipients in Singapore may contact the Singapore branch of Barclays Bank PLC, whose registered address is OneRaffles Quay Level 28, South Tower, Singapore 048583.

Barclays Bank PLC, Australia Branch (ARBN 062 449 585, AFSL 246617) is distributing this material in Australia. It is directed at 'wholesale clients' as definedby Australian Corporations Act 2001.

IRS Circular 230 Prepared Materials Disclaimer: Barclays does not provide tax advice and nothing contained herein should be construed to be tax advice.Please be advised that any discussion of U.S. tax matters contained herein (including any attachments) (i) is not intended or written to be used, and cannotbe used, by you for the purpose of avoiding U.S. tax-related penalties; and (ii) was written to support the promotion or marketing of the transactions or othermatters addressed herein. Accordingly, you should seek advice based on your particular circumstances from an independent tax advisor.

© Copyright Barclays Bank PLC (2013). All rights reserved. No part of this publication may be reproduced in any manner without the prior writtenpermission of Barclays. Barclays Bank PLC is registered in England No. 1026167. Registered office 1 Churchill Place, London, E14 5HP. Additional informationregarding this publication will be furnished upon request.

US08-000001

The VIX Compass: August 28xa.yimg.com/.../113602213/name/Barclays_The_VIX_Compass_Augu… ·...

Documents

Transcript of The VIX Compass: August 28xa.yimg.com/.../113602213/name/Barclays_The_VIX_Compass_Augu… ·...