V O L AT I L I T Y A S A N A S S E T C L A S S

By Cliff Stanton CFA, Chief Investment Officer, Prima Capital

The Potential of the VIX as a Hedging Tool and the Shortcomings of VIX Exchange Traded Notes

the only thing we have to fear is fear itself. Franklin D. Roosevelt That famous phrase was uttered in the midst of a banking crisis in 1933; a situation not entirely unlike the one investors faced in 2008. A recounting of the events of 2008 would be highly redundant at this point, but the experience of 2008 allows us to twist President Roosevelts statement into an interesting question. If the only thing to fear is fear itself, can fear in turn be a good defense against fear? Enter the VIX. The Chicago Board Options Exchange (CBOE) Market Volatility Index, or VIX, was conceived in 1993 by Professor Robert E. Whaley of Duke University to provide a benchmark of expected short term volatility. According to the CBOE, VIX measures 30-day expected volatility of the S&P 500 Index. The components of VIX are nearand next-term put and call options, usually in the first and second S&P 500 Index (SPX) contract months. As we will demonstrate, the value of volatility itself lies in the fact that it is negatively correlated to the returns of the equity market, and becomes increasingly so as market declines accelerate. As a result, long exposure to volatility could provide increasing levels of portfolio protection exactly when investors are most in need of such protection. Another way to think about this is that because most investors are net long equities, they are implicitly short volatility, and therefore hedging that exposure may be prudent. The theoretical knowledge and empirical understanding of implied volatility, as defined by the VIX, was put into play in 2004 when VIX futures began trading, followed in 2006 by the creation of options on the VIX1. More recently, in 2009 Barclays Capital launched exchange traded notes that track VIX futures contracts2. The development of exchange traded notes (ETNs) on the VIX has special relevance for the private wealth management industry. While institutional investors regularly utilize derivative instruments, the ability of financial advisors to use futures, options or swaps is often greatly limited, if not precluded, when working with high net worth individuals. The reasons vary, but include a host of operational considerations, a lack of familiarity with derivatives at the advisor level, and/or reluctance of individual investors to allocate capital to

Prior to 2004 options strategies and variance swaps had been utilized to capture exposure to volatility.

Symbols: VXX and VXZ

Prima Capital. | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 2/17

exotic structures. The availability of an exchange traded note solves these problems, although as we will see, not without introducing other issues. Our evaluation of these ETNs and the volatility index on which they are based attempted to answer the following questions: 1. 2. 3. What is the nature of equity volatility? What are the characteristics of the VIX and does it adequately capture true volatility? Are the VIX futures and related ETNs in turn able to harness the qualities of the VIX, or is something lost in translation? 4. What would an allocation to volatility as an asset class bring to a diversified portfolio?

But before we begin to address these questions, we must delineate the primary difference between volatility and other asset classes. Unlike a share of stock, a bond, or an ounce of gold, volatility has no intrinsic value. As such, over time the expected return to this mean reverting measure will be equal to the cost of attaining exposure. In other words, long term exposure to volatility is simply a cost, a drag on portfolio performance. For the truly long term investor, who isnt negatively impacted by short term volatility, there is no reason to bear such a cost, but for the rest of us, volatility may prove useful. What is the nature of equity volatility? The following chart demonstrates a few interesting features of volatility. For one, the downside is truncated. Unlike other assets, volatility cannot go to zero, and over the past three decades, it has spent little time below 10%. In addition, volatility is not normally distributed; it has exceedingly fat tails, as indicated by high kurtosis (i.e. peakedness of the distribution). Finally, it is itself highly volatile and positively skewed, making long
S&P500:Actual 30DayAnnualized Volatility
(source:S&P)

100.00% 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00%
Feb50 Feb53 Feb56 Feb59 Feb62 Feb65 Feb68 Feb71 Feb74 Feb77 Feb80 Feb83 Feb86 Feb89 Feb92 Feb95 Feb98 Feb01 Feb04 Feb07 Feb10

Mean=13.2% Median=11.4% Maximum=90.8% Minimum=2.7% Skewness=3.9 Kurtosis=26.1 Volatility=124.9%(annualized)

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 3/17

exposure to volatility quite beneficial to investors. As indicated, the prior chart shows actual 30-day volatility, selected in order to correspond to 30-day implied volatility as measured by the VIX. Additionally, a review of the return distribution of the S&P 500 is informative. From the chart below we see that daily returns for the S&P 500 are also non-normal, that is, fat tails exist (even if the scale prevents us from seeing them). But unlike 30-day volatility, this distribution is negatively skewed. Further, the maximum daily loss has been roughly twice the size of the maximum daily gain; something to keep in mind as we take a look at the VIX in the next section of this paper.

S&P500:Daily %ChangeJan.1950 Feb.2010

(source:Standard&Poors)

4500 4000 3500 3000 Maximum=11.6% Minimum=20.5% Skewness=0.7 Kurtosis=22.8 Volatility=15.3%(annualized)

#ofObservations

2500 2000 1500 1000 500 0

What are the characteristics of the VIX and does it adequately capture true volatility? It is impossible to invest in volatility directly, and therefore, the prior analysis of the volatility of the S&P 500 is helpful as a backdrop, but not particularly useful in practice. The VIX gets us one step closer to investing in volatility, but in and of itself the VIX suffers from the same problem, one cannot invest directly in the VIX; it is not a tradable index. Before exploring products that attempt to generate returns dependent on the VIX, we need to determine whether the VIX itself does a good job of capturing the positive hedging characteristics of actual volatility. To do that, we should consider the transmission mechanism by which equity volatility impacts the VIX. Whaley (2008) and Siegel (2007) both touch on this mechanism. And that is, when equity markets are falling, the demand for both out-of-the-money and at-the-money put options increases, driving up the price of those options, and concurrently, implied volatility. Siegel also points out that arbitrageurs who sell puts, must also sell stocks in order to hedge their position and remain delta neutral. The more puts they sell, the more stocks they sell, which may exacerbate a selloff and increase the negative correlation between implied volatility and equities.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 4/17

The following charts show the closing level and the daily return distribution, respectively, of the VIX going back to 1990. As is the case with actual volatility, the series is mean reverting, the downside is truncated, the distribution is non-normal and it is positively skewed.

CBOEVolatility Index(VIX)
(DailyClosingLevels)

90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00

Mean=20.3 Median=18.8 Maximum=80.9 Minimum=9.3

Jul90

Jul91

Jul92

Jul93

Jul94

Jul95

Jul96

Jul97

Jul98

Jul99

Jul00

Jul01

Jul02

Jul03

Jul04

Jul05

Jul06

Jul07

Jul08

CBOESpotVIX:Daily%ChangeJan.1990 Feb.2010

300 Maximum=64.2% Minimum=25.9% Skewness=1.22 Kurtosis=7.54 Volatility=95.9%(annualized)

250

200

150

100

50

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Jul09

Volatility as an Asset Class

Page 5/17

While considering the return distribution of the daily changes of the VIX, recall the return profile of the S&P 500 that was presented earlier3. The best single day for the S&P 500 going back to 1950 was +11.6%, and its worst one day loss, suffered in October of 1987 on Black Monday, was -20.5%. While the worst one day loss on the VIX is similar in magnitude at -25.9%, the best single day for the VIX was +64.2%. The return profile for the VIX is indeed asymmetrical, and that asymmetry can be demonstrated by comparing 30-day returns on the S&P 500 to 30-day returns on the VIX.

Monthly%ChangeintheS&P500vs.VIXJan.1990 Feb.2010
200.0%

150.0%

100.0%

VIXMonthly%Chg

50.0%

0.0% 40.0% 30.0% 20.0% 10.0% 0.0% 50.0% 10.0% 20.0% 30.0%

100.0% S&P500Monthly%Chg

The slope of the line is steepest when the S&P 500 is deep in the red, but flattens out as equities post positive returns. What does this tell us? When it is most needed, that is, when the equity markets are falling dramatically, the delta, or rate of change on the VIX is the greatest. This type of profile allows for a small commitment of capital to provide meaningful protection to a portfolio, which we will explore later. But lets continue with the question of whether the VIX captures actual volatility in a meaningful way. As option traders are acutely aware, implied volatility is almost always higher than actual subsequent volatility, up until the moment when actual volatility spikes. It is for this reason that selling naked options has been likened to picking up nickels in front of a steam roller it works and its easy right up until the moment when you get run over. Small winsmall winsmall winbig loss, is not an attractive return profile for most investors.

Please note that the time periods of these two return distributions differ, as the S&P 500 data goes back

to 1950 whereas the VIX data begins in 1990.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 6/17

S&P500:Impliedvs.Actual Subsequent Volatility

(source:S&P,CBOEandPrimaCapital)

90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00

Jan90

Jan91

Jan92

Jan93

Jan94

Jan95

Jan96

Jan97

Jan98

Jan99

Jan00

Jan01

Jan02

Jan03

Jan04

Jan05

Jan06

Jan07

Jan08

Jan09

VIX

Actual30DayVol

While implied volatility is almost always higher than actual subsequent 30-day volatility (see above), the correlation between the two series is high at 0.73. But if we are considering using VIX exposure as a hedge, we arent particularly concerned with the ability of the VIX to forecast future volatility (and thats not a question that a simple correlation analysis can answer in any event). What we are concerned with is both the correlation and sensitivity between the daily changes in the S&P 500 and the daily changes in the VIX. By regressing the VIX on the price changes of the S&P 500 we calculated the following statistics4 on both a daily and monthly basis.

S&P 500vs.

Correlation

Bull Correlation

Bear Correlation

Beta

Bull Beta

Bear Beta

Up Capture

Jan10

Down Capture

VIX (daily)

-0.7

-0.4

-0.6

-3.5

-2.2

-3.9

-388%

-434%

VIX (monthly)

-0.6

-0.2

-0.5

-2.6

-0.9

-3.0

-215%

-361%

The time period over which the calculations were made is from January 1990 through February 2010.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 7/17

As you can see, these statistics confirm what weve already been discussing, i.e. that the return profile to the VIX is quite different in up markets than it is in down markets. The correlation statistics become more negative during down markets and the beta statistics increase, and the down market capture ratios are greater than the up market capture ratios. These relationships are exactly what wed like to see in a hedging vehicle. There is one final point that wed like to make regarding the VIX and its mean reverting ways. Our regression analysis revealed that the strongest linkage between the starting point on the VIX and subsequent returns was around the five year mark. Meaning that if an investor was able to invest in spot VIX, and held that position for five years, the return to that investor would be largely dictated by the beginning level of volatility. This can be thought of in similar terms to the relationship between the starting valuation level for the equity market and subsequent equity returns. To demonstrate this point, we constructed the following table, indicating the level of the VIX at that time a hypothetical investment would have been made, and the subsequent 5-year return on that investment.

SPX VIX Closing Level at Time of Investment

Average Subsequent 5-Yr Annualized Return

<11.96

16.0%

11.96-16.96

10.9%

16.96-21.96

0.3%

21.96-26.96

-7.9%

26.96-31.96

-10.5%

>31.96

-11.0%

The takeaway from this is that the payoff to volatility is greatest when it is in effect cheapest, that is, when market participants are complacent and risk seeking. Once the markets are in the midst of a selloff and market participants are scared stiff, the expected payoff to VIX exposure at that point turns negative. Yet another instance where being contrarian pays off.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 8/17

Are the VIX futures and related ETNs in turn able to harness the qualities of the VIX, or is something lost in translation? Thus far weve been able to make the case that volatility as an asset class has attractive features that could allow it to be an effective hedge against negative equity outcomes. Further, weve shown that while the VIX is intended to reflect expectations of future near term volatility, it does more than an adequate job of capturing changes to actual volatility as those changes occur. The question remaining is whether futures based on the VIX, and in turn the ETNs, will convey similar advantages to investors. One problem at this point in time is the lack of history on which to make a meaningful analysis. The Barclays ETNs were launched in January of 2009, providing little more than a year of data. Even so, the ETNs are based on futures contracts and futures have been trading on the VIX since late 2004. As such, we begin with an analysis of those futures contracts. Like all futures contracts, a term structure exists for VIX futures. That term structure is typically in contango5 when volatility is stable or low, and moves to backwardation6 when volatility is high. The term structure of the VIX as of March 31, 2010 is shown below7, a time when volatility was subdued and the futures were in contango. During such periods, rolling the futures contracts results in a significantly negative roll yield. Specifically, buying higher priced, longer dated futures contracts and selling them at lower prices as they approach maturity, rolling the proceeds into yet another longer dated futures contract, results in a loss.
VIXFuturesTermStructureasofMarch31,2010
30

25

20

15

10

0
Apr10 Aug10 Apr11 Aug11 Apr12 Aug12 Feb11 Feb12 Jun10 Jun11 Dec10 Dec11 Jun12 Dec12 Oct10 Oct11 Oct12

Contango is a term used in the futures market to describe an upward sloping forward curve.

Backwardation is a term used in the futures market to describe a downward sloping forward curve.

Source: CBOE

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 9/17

However, in times of high volatility, such as that witnessed in October of 2008, the term structure will likely be in backwardation. As such there will be a positive roll yield earned by investors, as contracts are rolled forward.

VIXFuturesTermStructureasofOctober15,2008
70

60

50

40

30

20

10

0
Aug09 Aug10 Jan10 Apr09 Apr10 Jan09 Nov09 Nov08 May09 May10 Nov10 Feb09 Sep09 Feb10 Sep10 Jul09 Jun09 Jun10 Jul10 Dec09 Mar09 Mar10 Dec08 Dec10 Oct09 Oct10

This concept of roll yield adds to the story developed earlier about the starting level on the VIX and subsequent returns, and these two aspects work against one another. In times of high volatility, the roll yield is positive, but the expected return on the VIX turns negative. In contrast, when volatility is low, the expected return is positive, but the roll yield is negative, and maintaining such a position becomes expensive. To explore what that cost might look like over time we considered a hypothetical portfolio consisting of spot VIX versus a portfolio consisting of VIX futures contracts. Barclays Capital provided an index series representing a hypothetical VIX futures portfolio managed on the same basis as the iPath S&P 500 VIX Short Term Futures ETN (ticker: VXX; gross of fees). The following chart shows the value of these two hypothetical portfolios.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 10/17

Growthof$100:SpotVIXvs.VIXFuturesPortfolio
800

700

600

500

400

300

200

100

0
Feb06 Feb07 Feb08 Feb09 Dec05 Dec06 Dec07 Dec08 Aug06 Aug07 Aug08 Aug09 Dec09 Feb10 Jun06 Jun07 Jun08 Oct06 Oct07 Oct08 Jun09 Apr06 Apr07 Apr08 Apr09 Oct09 Apr10

HypotheticalFuturesPortfolio

SpotVIX

It is at this point that our story, which had been quite promising, starts to fall apart. As a reminder, an investor cannot invest in spot VIX, so the theoretical portfolio value indicated above is merely a reference against which we can judge the efficacy of a futures portfolio in capturing the theoretical returns to spot VIX. Having said that, a constant futures position fails miserably in delivering the returns produced by spot VIX. Over the time period indicated, the annualized return to spot VIX was almost 17%, whereas the futures portfolio generated an annualized return of approximately -22%. That differential, almost 40% annualized, is primarily due to the roll yield. It is also impacted by a difference in the time periods utilized in calculating the two series, as the VIX reflects implied volatility in the near term and next term options contracts, while the futures on the VIX are priced off of future expectations for implied volatility. As an example, a 3-month VIX futures contract is measuring what the expected implied volatility on the S&P 500 Index will be in three months time. This time discrepancy is the reason that VIX futures exhibit lower volatility than spot VIX. In defense of the cost of rolling futures contracts, Barclays points out that roll yield can be likened to the theta (time decay) of put options, and claims that the roll yield is lower than the implied realized premium in put options. Further, because of the term structure of VIX futures, Barclays offers two different ETNs. The short term VIX ETN (VXX) is based on the 1 and 2 month futures contracts, whereas the medium term VIX ETN (VXZ) is based on the 4-7 month futures contracts, both of which are managed with a daily roll methodology. Because the mid range of the term structure is flatter, roll costs are minimized, and in a low volatility environment when the roll yield is in contango, investors are better off in VXZ. The downside to VXZ is lower beta relative to VIX. Barclays calculates that the highest beta of VIX futures relative to actual VIX is in the 1month and 2-month contracts, and turns out to be about 0.5 and 0.35, respectively. That beta decreases with maturity, and so VXX is best when volatility spikes.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 11/17

Due to the issues explored above, the return profile for VXX and VXZ will differ materially from the VIX. Consider the daily change in the price of VXX relative to the daily change in VIX since inception of the ETNs, graphed below.

VXXiPathETNvs.VIX:DailyPriceChangeDifferential
25.00%

20.00%

15.00%

10.00%

5.00%

0.00%

5.00%

10.00%
10/31/2009 11/30/2009 12/31/2009 1/30/2009 2/28/2009 3/31/2009 4/30/2009 5/31/2009 6/30/2009 7/31/2009 8/31/2009 9/30/2009 1/31/2010 2/28/2010 3/31/2010 4/30/2010

As you can see, the daily differential is typically large, and in fact the average differential is 2.5% on an absolute basis. Further, as discussed previously, the beta of VXX to VIX is far below 1.0. Our calculation of the beta of VXX to VIX on a daily basis is 0.47. Even so, VXX is highly correlated to VIX; the correlation between the daily price changes is around 0.86. What can we conclude from this analysis? It is clear that while the VIX itself has attractive characteristics that make it a highly effective hedging tool, the futures contracts, and ETNs in turn, are far less compelling for investors wishing to take a buy and hold approach to what is effectively portfolio insurance. Holding the position long term not only results in a direct cost of 89 basis points for the ETNs, but more importantly, a materially negative roll cost. Our first thought as to how to deal with this issue was to come up with a simple trading strategy based on the level of the VIX when a position was initiated. For example, one could take a full position in VXX or VXZ when volatility was near 1-standard deviation below the mean, virtually ensuring a positive expected return to volatility given its mean reverting tendencies. That position could be pared as VIX reached its mean, and then exited at 1-standard deviation above the mean, at a time when expected returns are negative. However, as discussed earlier, the term structure of the VIX results in the highest roll costs during periods of low volatility, and while volatility is mean reverting, it can take years for that to occur, all the while placing a significant drag on an investors portfolio. More intricate trading models based on much shorter time frames could work,
Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 12/17

although that detracts significantly from the use of these types of products within the private wealth management industry. Alternatively, positions could be added whenever an advisor believes that the likelihood of a volatility spike is increasing, such as around major macro events, and that would be highly effective, but unfortunately, not all major market crises have an observable ramp-up ex-ante. Advisors wishing to utilize these ETNs are necessarily going to be injecting significant timing risk into their use. Ultimately, we do not believe that they are appropriate for buy and hold investors, even though they do react quite nicely when truly needed.

What would an allocation to volatility as an asset class bring to a diversified portfolio? While weve concluded that an investment in the VIX through ETNs is too costly to be practical as a strategic allocation to a portfolio, we decided to explore what an investment in spot VIX could add to a portfolio, if that could be achieved. Ours is an ever changing industry and more optimal ways of harnessing the VIX may come along. Every asset added to a portfolio should serve at least one of two purposes, return enhancement or risk reduction. Those descriptors are relative, meaning that the role that an asset plays is relative to another asset. Strategies employing hedging techniques more often than not provide risk reduction to an equity heavy portfolio, but typically provide return enhancement as well to a fixed income heavy portfolio. Many strategies employed by hedge funds (and increasingly hedge fund like mutual funds) attempt to reduce systematic risk (e.g. equity risk, credit risk, interest rate risk) within their own portfolio, but do not directly offset those very same risks contained elsewhere in an investors portfolio. For example, an equity market neutral fund may not introduce any equity beta into a portfolio, but it doesnt offset the equity beta that comes from a large cap growth manager in the portfolio. In order to protect a portfolio against large losses in times of market stress, an investor needs a direct hedge, i.e. an asset that is negatively correlated to other assets in the portfolio. Volatility, as mentioned previously, is just such an asset. We constructed a series of efficient frontiers using forward looking return and risk expectations. Our goal is to be as realistic as possible about the risk and return characteristics of this hypothetical VIX position, and we understand that exposure to any asset comes with a cost. In this case, for simplicity sake, we assumed that the return expectation to the VIX was -0.89%, which mirrors the explicit cost to the VIX ETNs (but importantly, ignores the roll cost described earlier). Even with a negative expected return, inclusion of the VIX improves efficiency. What you can see below is that an efficient frontier constructed solely with a basic set of asset classes, i.e. domestic stocks, bonds and cash, can be improved by adding a VIX position. That improved upon efficient frontier pushes out even farther when constructed using a complete asset class set (which includes a variety of assets from domestic stocks, to emerging market debt to bank loans, in addition to the VIX). You can also see that a small allocation to the VIX can have a big impact. The portfolio labeled BG&I, which is a balanced growth and income portfolio, has an expected return of 6.85% with expected volatility of 10.5%. A 3.5% allocation to the VIX at the expense of the equity exposure (i.e. BG&I + 3.5% VIX) decreases expected return by 30 basis points, but decreases risk by 216 basis points; a valuable tradeoff.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 13/17

TheEfficientFrontier
(Calculatedwith5YearForwardLookingEstimates) 12.0%

10.0%

8.0%

ExpectedReturn

6.0% 60%Equity/40%Bonds 4.0% 55%Equity/40%Bonds/5%VIX 50%Equity/40%Bonds/10%VIX 2.0%

0.0% 0.0% 5.0% 10.0% 15.0% Risk

CompleteAssetClassSet BasicAssetClassSet BasicAssetClassSet+VIX BG&I NavePortfolios BG&I+3.5%VIX

20.0%

25.0%

30.0%

35.0%

We also looked at what inclusion of the VIX would mean to Value at Risk8 and loss probability, comparing a basic 60/40 portfolio with the same BG&I and BG&I + 3.5% VIX portfolios identified on the efficient frontier.

ValueatRiskfora$1Million PortfolioOvera1YearHorizon

$146,935 $160,000.00 $140,000.00 $109,381 $120,000.00 $100,000.00 $66,376 $80,000.00 $60,000.00 $40,000.00 $20,000.00 $ EndofHorizonValueatRisk(5%)
60/40 BG&I

$134,280

$101,587 $95,282

WithinHorizonValueatRisk(5%)
BG&I+3.5%VIX

VaR The maximum likely loss over the indicated time period at the 95% confidence level.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 14/17

Probability ofaLossof5%orGreaterOvera1YearHorizon

45.7% 50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% EndofHorizonValueatRisk(5%)
60/40 BG&I

40.7%

27.7%

15.5% 12.6% 7.7%

WithinHorizonValueatRisk(5%)
BG&I+3.5%VIX

Again, inclusion of the VIX, with only a small allocation of capital, has a noticeable impact. What is most striking is the decrease in the probability of a loss of 5% or greater over a 1-year horizon (on a continuous basis). A 60/40 portfolio has a 46% chance of a loss of 5% or more, whereas the BG&I + 3.5% VIX has only a 28% chance of such a loss.

Conclusion Volatility as an asset class is compelling. It is negatively correlated to the returns to the equity markets, and becomes increasingly so as market declines accelerate. As a result, long exposure to volatility could provide increasing levels of portfolio protection exactly when investors are most in need of such protection. The convexity provided by volatility means that a relatively small allocation of capital can provide meaningful protection. However, as we demonstrated, the benefit to volatility erodes as we transfer its characteristics from actual volatility, to implied volatility, to futures on implied volatility, and finally, to ETNs based on the futures on implied volatility. In the end, what appeared to be a powerful diversifying tool for strategic asset allocators, ends up being appropriate only to express shorter term views on volatility. What other options does an investor have? The use of dedicated short managers can certainly be effective in mitigating total equity risk, and many hedge fund of fund managers do just that. However, the return profile tends to be more linear with the return to the equity market. That is, a 4% decline in the S&P 500 would likely result in a return of 4% to a dedicated short manager with at least enough skill to offset fees. But the asymmetric return profile of the VIX cannot be matched by short managers.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 15/17

Another possibility is of course equity options. But Szado (2009) demonstrated that VIX calls provide a more efficient means of diversification than S&P puts. Further, Jacob and Rasiel (2009) made the following observations about buying options: Options have exposure not just to volatility, but also to market direction. To maintain directional neutrality requires frequent updating of a delta hedge position on the underlying asset. Options are a wasting assetTherefore, options cannot be included as an asset in a buy-and-hold portfolio. Finally, what about hedge funds in general? We advocate the inclusion of hedging strategies in a portfolio, to complement long-only equity and fixed income managers. However, as mentioned previously, while many strategies employed by hedge funds attempt to reduce systematic risk (e.g. equity risk, credit risk, interest rate risk) within their own portfolios, that risk reduction does not directly offset those very same risks contained elsewhere in an investors portfolio. Equity beta coming from the large cap growth manager is not offset by the hedged equity position within a convertible arbitrage strategy. Given our findings regarding the VIX ETNs, as well as the limitations of other potential investments as true hedges against market crises, we are forced to conclude that the search for the holy grail of hedging for strategically allocated portfolios must continue. In the absence of a perfect hedging vehicle, investors should continue to rely on tried and true measures of risk mitigation. First and foremost, know what you are investing in and why; thoroughly evaluate the risk and return characteristics of each investment, and properly balance those characteristics in order to achieve the desired results. Be realistic in forecasting return expectations, and recognize that risk management isnt simply about the probability of being wrong, but the magnitude of the loss if you are wrong. Second, diversify broadly. Go beyond the nave approach, which can be greatly improved upon through proper portfolio construction in which asset allocation and manager selection are utilized as separate and distinct tools to solve different problems. Investors utilizing these principals have been well served over time (2008 notwithstanding), and will likely reap the rewards of doing so in the future.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 16/17

References Barclays Capital. 2009. Trading Equity Volatility with S&P 500 VIX Short Term & Mid-Term Futures Indices. Institutional Investor Presentation Darnell, Max. 2009. What Volatility Tells Us about Diversification and Risk Management. CFA Institute Conference Proceeding: Asset and Risk Allocation (Philadelphia 2009) Deshpande, Maneesh and Rohit Bhatia. 2010. Understanding VIX ETNs. Barclays Capital Conference Call Presentation Evans, Mark and Raymond Chan. 2010. Managing the Left Tail: A Framework for Market Risk Management. Goldman Sachs Asset Management Conference Call Presentation Goldstein, Daniel G. and Nassim Nicholas Taleb. 2007. We Don't Quite Know What We are Talking About When We Talk About Volatility. The Journal of Portfolio Management, vol. 33, no. 4: 84-86. Jacob, Jai and Emma Rasiel, PhD. 2009. Index Volatility Futures in Asset Allocation: A Hedging Framework. http://www.lazardnet.com/lam/global/investment_research.html Kritzman, Mark. 2009. Managing Assets in Turbulent Markets. CFA Institute Conference Proceeding: Asset Allocation for Private Clients (Atlanta 2008) Lustig, Michael. 2007. Using Derivatives to Enhance Returns and Manage Risk. CFA Institute Conference Proceeding: Fixed Income Management (San Francisco 2006) Moran, Matthew T. 2004. Taking a Ride on the Volatile Side. The Journal of Indexes, October/November 2004 Moran, Matthew T. and Srikant Dash. 2007. "VIX Futures and Options: Pricing and Using Volatility Products to Manage Downside Risk and Improve Efficiency in Equity Portfolios." The Journal of Trading, Summer 2007: 96-105. Nossman, Marcus and Anders Wilhelmsson. 2009. Is the VIX Futures Market Able to Predict the VIX Index? A Test of the Expectation Hypothesis. The Journal of Alternative Investments, Fall 2009: 54-67. Siegel, Jeremy J. 2007. Stocks for the Long Run. 4th Edition. New York: McGraw-Hill Szado, Edward. 2009. VIX Futures and Options A Case Study of Portfolio Diversification During the 2008 Financial Crisis. The Journal of Alternative Investments, Fall 2009: 68 - 85. Whaley, Robert E. 2008. Understanding VIX. Available at SSRN: http://ssrn.com/abstract=1296743

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com

Volatility as an Asset Class

Page 17/17

About Prima Capital

Prima Capital, a Matrix Financial Solutions affiliate company, is a leading provider of due diligence, research applications, asset allocation modeling, and multi-manager portfolios to the wealth management and retirement industries. For over a decade, Prima Capital has provided objective due diligence and powerful evaluation tools to advisors and wealth management firms of all types and sizes. This combination of advice and technology helps empower advisors and improve the efficiencies and distinction of wealth management programs. Today, Prima Capital is an independent source for information on managed account strategies (UMA and SMA), mutual funds, ETFs, and alternative investments through the PrimaGuide research application and the Prima AllocationGuideTM proposal system. Prima Capitals Target Risk Portfolios bundle the best managers from Primas unbiased research into a series of globally diversified, professionally constructed portfolios. Through an affiliation with Matrix Financial Solutions, Prima Capital advances open architecture retirement solutions through the MG Trust RetireFocus Funds (target-date collective investment funds) and the RetireTool(k)itTM application. For more information, visit www.primacapital.com.

Intended for investment professional use only. This commentary is not intended as an offer or solicitation for the purchase or sale of any security or other financial instrument. It is based upon information that Prima Capital Holding, Inc. (Prima Capital) considers to be reliable but we do not warrant its accuracy or completeness. Assumptions, opinions and estimates constitute our judgment as of the date of this material and are subject to change without notice. Neither Prima Capital nor its affiliates are responsible for any errors or omissions or for results obtained from the use of this information.

Prima Capital, Prima Guide and the Prima Capital logo are trademarks or registered trademarks of Prima Capital Holding, Inc. All other trademarks used herein are the property of their respective owners. Produced exclusively for the use of our clients and prospective clients. This document may not be reproduced in whole or in part without the express, written permission of Prima Capital.

Prima Capital | 303.573.7315 | 303.573.7362 fax | www.primacapital.com